Author here! Thanks for all the comments. This post isn't meant as a lesson or a judgment, something to tell "just shut up and adapt like my ancestor did". The situation is hard, and I wasn't trying to dismiss anyone's anxiety. No one wants to hear "you'll be fine" when something bad just happened.
It's mostly a personal story. You can see it as a kind of tribute to a great-great-grandfather I never met, partly because I'd never heard a story like it before. It's banal but still a big adventure in the family. I find it fascinating that he grew up in a completely different world from mine and still felt some of the same doubts and anxieties I do today. And I'm really proud of him, because his "pivot" let my family stay in the village instead of maybe leaving to work in factories in the city. My father loved his job, and in a way that's thanks to him.
I appreciate your wholesome message, but I feel a bit is overlooked.
Horses and engines are in a tangible, corporeal domain. I almost feel like pivoting could be easier with that: the tools are still the hands. Also, pivoting might be easier especially when no one is in that space of repairing cars in a small village at the time. Low competition. I also think time allowed to pivot is much higher: automobiles were expensive and time to adoption is higher, I wonder how long it took to be able to make a living?
Data, numbers, programming, this is in a ephemeral domain. Competition is high, there are a lot of developers. Time to adoption is extremely fast. In the automobile era, someone may have had five to ten years to pivot. Now? Maybe a year?
Is it about making things? Maybe, but pivoting and making a career of it is not a straight path, and sometimes people get lucky in choosing the path.
>> Many blacksmiths added the mechanical skills needed to service and repair automobiles, particularly in smaller towns and rural areas. In short order, some of the more commonly replaced automobile parts, such as tires, were being stocked by blacksmith’s, others which were necessary to a vehicle’s repair were ordered by telegraph or telephone for delivery to the blacksmith’s shop. Blacksmiths and service stations soon had competition from the automobile parts store, another new specialty which blossomed along the popularity of traveling by car.
Comparing programmers to farriers is definitely strange on the surface. However I think it's quite unfair to reject the authors premise due to the ephemeral nature of some software. And I think it's downright arrogant to assume moving from horses to cars wasn't seriously challenging for his ancestor. But yeah, I would love to chat with the old man about what that transition looked like.
Also, AI adoption might look like it's moving fast for individuals, however I see many companies struggling to actually utilise it effectively. Business processes are slow moving political machines. Sure a large part of the software industry is ephemeral, indirectly providing capital value. However many jobs are still very tangible; embedded, backend web systems, networking etc etc. Instability in these systems absolutely results in tangible loss of value.
It will absolutely take decades before we figure out 'correct' AI usage at an industry scale. Heck, even frontier labs are still undecided about best practices.
> Comparing programmers to farriers is definitely strange on the surface.
Just for another datapoint: two of my great-great-grandfathers in two family branches, in the same small village in France, were respectively farrier and black-smith, while I'm a software engineer. One son of the black-smith continued the trade as a side activity of being a plumber. So I had to think about how my career choice (and those of my siblings, uncles, cousins) relates to theirs.
My own perspective on those trades is that transforming metal was the top engineering scale in the countryside in their era. In those times (XIXth century, early XXth), learning the job required to go work with different masters around France ("tour de France" had another meaning before cycling). While it was still manual work, it was highly skilled job which required both technical skills and creativity, which are also key skills for software engineering.
Another thing is that, unlike AI, automobiles were (somewhat kinda) going to "replace horses".
The AI situation is similar to someone saying "see, we're gonna keep the horses exactly as they are, but replace farriers, trainers, grooms, stable hands, saddlers, vets, and coachmen with absolutely nobody other than some manager type".
The AI industry can't really say they're gonna replace SaaS, because SaaS is the lifeline of their business.
So they just claim they're gonna replace several classes of professionals.
The total market addressed by current-shaped LLMs is not "software development" - it's approximately all of white-collar jobs. And with recent advancements in robotics, blue-collar work isn't really much safer. Hardware is getting better, and the one thing that was stopping robotics from entering our own workplaces and living spaces, the ability to make decisions on the fly in heterogenous, unpredictable environment - the judgement - is literally what LLMs themselves are embodying.
So I agree, we all likely still have 3-5 years to figure this out - but also no indication where to go, or if there even exists a direction to go in.
art and music for some, but personally I'm most interested in seeing where science goes. Reproducibility of experiments is gonna be first. How many experiments don't reproduce, or have unwritten biases in the papers. Room temperature semiconductors. Fusion seems actually a decade away, finally. Research into better treatments for diseases, how to help people be smarter, live their best life. There's plenty more to do now that making CRUD apps is so much easier. Maybe AI will become creative enough to not need humans for that, but it'll still need someone to drive it and choose which direction it goes.
Art and music? Ok for live performance. But recorded performance is directly in competion with genAI for revenue streams, with content used without permission for more genAI training.
It can't though. You still need to run real trials on living humans, living their lives and taking experimental drugs/procedures/interventions and running statistics on them.
AI can do everything you described except the "living out a human life" part - but it can iterate on ever better simulations of that, or worse, it may reward-hack and... well, I'd rather not be the human in that story.
Do you really think humans are going to be the users? Sure for tomorrow and next month, but the clock is ticking. I got mad at our internal CS peeps the other day. They were demanding an internal tool. I was like we went through hell this summer replacing the front end so it's agent usable, just get Claude to do that. They were like oh...
Sounds amazing. How on earth did he go from shoeing horses to learning about the internal combustion engine? Seems quite the leap that not everyone would be capable of.
Joke aside, it was not shoeing horses but also working on the carts, etc. Probably bikes too. So at least he reused some mechanical skills, and probably tools, he already had. Only the engine was really new.
And many people living in small villages back then had basic knowledge in many disciplines, including electricity. When you had a farm for instance, you needed to be able to repair yourself anything.
This happened to my grandfather (blacksmith to working on cars and motorcycles), who was illiterate. From what I gather, it was a lifetime of trial-and-error-and-keep-what-works.
Early cars were called horseless carriages for a reason.
Car bodies were carriages for few first decades of car history. So if you can fix and maintain horse carriage - you can maintain car carriage.
Engines were simple too and designed to be maintained in a field environment with few tools. So it’s reasonable that this skill was learned, especially by mechanically inclined people.
Early automobiles were very simple machines. I you knew how to repair a carriage or a wagon, you were pretty close to being able to work on an early automobile. You should check them out.
Conversely, carriages (and other pre-automobile machines) were also surprisingly complicated. Steering mechanisms, leaf springs, brakes, and a variety of linkages including the horse harnesses themselves (which had to account for the animal's gait).
Honestly I think people forget this fact. You don't have to go that far back to find vehicles that you absolutely could do anything to yourself with basic tools.
We're fixing up a 1987 Jeep Wrangler for our son. I could strip it down to the frame with 3 or 4 tools. It's kind of amazing how simple cars were, and that was in the late 80's.
My first car was a 1986 Grand Marquis and I spent every other weekend of high school fixing something on it. What an incredible waste of time: I worked a part time job so that I could afford to keep a car running so that I could drive between school, home, and my part time job instead of riding the bus.
But it was easy to work on. My dad (not a mechanic) usually had a strong hunch about what the problem was. Lights went dim while you were driving and five minutes later it sputtered out? Alternator is dead. Flip through the manual to find a diagram of where all the bolts are, take it off, run to Auto Zone, get another one, bolt it on. By the end it was usually dark and I was finishing the job with a shop light hung under the hood.
I drove a series of beaters from the 80s and 90s and did all my own repairs until my mid 20s. Alternators, water pumps, break pads, break lines, ignition solenoids. All of it is just parts that you unbolt and bolt a new one on. When I got my first modern car (a 2008!) one look under the hood made me realize I had joined the yuppie class and would not be changing my own spark plugs anymore.
I always did my own maintenance including changing spark plugs. Then I bought a Subaru. That weird 4 cylinder has plugs facing slightly down that you either have to cut the wheel well cowl or UNMOUNT AND JACK THE MOTOR OUT OF THE ENGINE BAY to change.
My solution was to use like 11 wobble extensions and pray that I didn't drop anything. 8 hours later I changed 4 plugs and understood why it cost $250 at the dealer to have them to it. Then I gladly started paying that money.
Yeah as long as you ignore the engine (ECU/PCM/TCUs) in cars in the 80s, they're quite simple. Go back to the early 70s-before and it's mostly all analog or trivial electronics.
This was the video that earned my sub. Even though it covers stuff I mostly already knew, it's done in such a way that that it breaks down the whys that may not be apparent to people without a background in this stuff.
This and the video on hybrids, are, imo important works, because they break things down in a way that disarms tribalism, and explain how it is that we got to where we are. There are very real stakes in ownership, right to repair, and freedom of movement that rest on people having a good baseline understanding of this stuff.
Even as late as the late 90s, probably into the mid 2000s for manuals. The complexity of vehicles from that era is mostly in understanding the role of the sensors (or that they even exist in the first place). Once you understand that, they go back to being pretty straightforward. I guess the exception here is probably immobilizers.
The complexity of modern vehicles comes down to software coupling between the systems, which (to my knowledge) started with automatic transmissions, and then spread to everything else. In very modern vehicles, even the sensors register with the ECU and communicate on encrypted channels.
The reason for these measures is ostensibly security and quality, but automakers don't seem to mind the fact that it also makes many repairs more difficult or impossible.
Shoeing horses properly is actually no joke and cars back then were laughably simple. I don’t see how any of that could feel relatable to people in software development now, when almost all jobs that require real smarts and competency require years of study and experience.
In the late 1800s there was a cottage industry of mechanics and ironmongers selling and repairing small steam and kerosene engines used by rural mills and farms. There were horse drawn steam threshing machines. Bicycle repair shops sprang up. The Wright brothers got into the bicycle business before horseless carriages. There were many intermediate tech skills that led up to working on motorcars.
Out of curiosity, what was the timeline ? Cause I'm afraid that, when you mentioned "I don't know what the other farriers" did, the answer may be "they were sent to the front and died by the 1000s".
Which starts to be my fear for the real "solution" to job displacement by AIs : people will just be sent to some far away conflict where they will transition to a career in "daffodil fertilization" :/
My father worked for the railroad as a blacksmith but he apprenticed to an old blacksmith who was a farrier before. There are still farriers to this day.
Some of us will need to move on to other things but there will probably still be a lot of farrier type jobs around.
Don’t worry about it. You’re just playing your part. The trite comparison to past successful innovations. The paragraph about the AI Revolution. The AI Inevitability Soothsaying (“Be willing to let go of the how to hold on to the why”).
The perfect HN frontpage material.
Of course I wrote the two preceding paragraphs before reading your webpage: “I am Co-founder and CTO of Tandem, an AI Assistant for SaaS apps.” Just another entrepenur with a psychological spin.
There is a quote from a CGP Grey video from a bit over a decade ago, "There isn't a rule of economics that says better technology makes more, better jobs for horses. It sounds shockingly dumb to even say that out loud, but swap horses for humans and suddenly people think it sounds about right." [1]
A couple hundred years ago, something like 70% of the population worked in agriculture. Technology replaced almost all of these jobs.
So far, we have reimagined old professions and created brand new ones. Is this ability of ours limitless?
This is one step away from walking into the same mistake all these discussions do, including replies to your comment.
Yes, we can probably invent new jobs and new career fields.
No, it does fuck all for people who lose their current jobs mid-way through their career or later, and watch themselves and their families be kicked down 2-3 rungs on the socioeconomic ladder, possibly throwing them into generational poverty.
Or, in short: yes, there will be new jobs. But not for you, and not for your children.
For the society and human civilizations, these techological revolutions were always a net positive down the line. For individuals caught in them, they rarely were.
That's why the Luddites were destroying looms. Not to fight the progress of technology, but to save their own families from being thrown into poverty.
Timeframe is also different, agriculture changed over generations and the kids that would not take over the farm had access to education and could move to cities. Now we may see a similar job destruction over a decade without alternatives available for much more people.
> had access to education and could move to cities
Or their rural economy collapsed, everyone was hungry and unemployed, and they moved somewhere new out of necessity and basically started over with nothing, living in squalid conditions and working very unsafe, unpleasant, and exploitative jobs.
I think the difference is the Luddites weren't happily designing and building innovative new looms for 50 years and now suddenly surprised-Pikachu that the automated-loom-maker has come for them.
What did you think "disruption" meant? Vibes, papers, new Javascript frameworks?
The amount of consumption is largely dependent on income level. There's a recent study from Moody's that says that in the US, the top 10% income bracket now accounts for roughly half of consumer spending. As wealth inequality continues to worsen, it makes more and more sense to solely target high earners because you make the same or more as selling higher amounts of lower priced product but with less overhead.
This doesn't make sense at all. It is a bad statistical analysis of consumer spending. [1] A few things from the thread on why by Levy on why it feels wrong .
First, why does it feel wrong?
1. the top 10% get about 50% of pre-tax income
2. they get 30 to 37% of disposable income after tax&transfers (depending on source)
3. and we know the rich have a higher savings rate, so consume a lesser share of their income (US, China)
May be not top 10% but top 0.1% are good at avoiding tax using various loopholes. Like borrowing against stocks (and other appreciating assets) till death to avoid paying CGT and debt is then repaid by heirs which do not pay CGT AFAIK. So they can spend more than income stats would suggest.
> As wealth inequality continues to worsen, it makes more and more sense to solely target high earners because you make the same or more as selling higher amounts of lower priced product but with less overhead.
This is exactly why everything like Disney has moved upmarket, with pricing rising far above inflation.
It works right up until some form of revolution, either politically or violently. We are far from anything like that but I'd rather not get close to the line. I think I'm a minority view though, despite being in the top 10% myself.
It is also shocking how little my life compares to the rich people I see. I don't own two houses or a yacht. We don't take yearly vacations to a chateau in France. The gap between the top 10% and the top 1% is insane. The gap between the 1% and the 0.1% is even larger still.
But hey... not like the rich getting richer and saying "screw you" to everyone else has ever led to negative consequences right? Nothing to see here, business as usual.
Caesar was a populist reformer. The Roman Republic was long destroyed at that point, due primarily to concentration of wealth and replacement of the plebians with slaves.
Crassus amasses so much wealth because Caesar helped him.
And Crassus was the rich man who was monopolizing all of Rome with his slave firefighters + real estate scheme. When the going got tough, they seekwd power from the rich guy, no one was stopping Crassus.
I think you have the wrong end of the stick (trying to map it onto contemporary politics?).
Caesar was killed by the same people (hardline conservative ultra-wealthy 'Optimate' families/factions) who also brutally murdered the tribune of the plebs (person appointed to beg the ultra-wealthy for favorable legislation) and his supporters not once but twice. He was killed for the same reason too - land redistribution, limits on slave labor. That was called 'tyranny' by the elites.
Crassus was wealthy before Caesar and died before Caesar took power. This kind of politicking was just a necessary evil in the Roman Republic - which always was an oligarch-controlled plutocracy and never a true democracy as we know it today (e.g. Athens). It was 'pay to play', sponsorship of the ultra-wealthy was not optional.
If by Disney you mean tickets to Disneyland and similar theme parks, then pricing rising above inflation actually reflects wealth inequality getting smaller, not bigger. A lot more people in the world than in decades past now have enough wealth to consider traveling to one of the small number of Disney theme parks in the world and spending time there, and there's only so many people who can fit in the park at once. The Disney corporation could (and has) built more parks around the world, but maybe you care about Disneyland specifically, and there's only one of those today just like in the 1950s when it was built.
The issue is that there isn't economic demand from that audience for approximately 90% of companies operating today. Maybe Disney can succeed by going upmarket, but can all the animation/filmmaking/TV/theme park companies they compete with?
The number of millionaires/billionaires/etc feels low enough that there won't be demand for dozens of companies in every possible industry. The ones that have the biggest fanbase/cultural footprint maybe stick around, but the majority would presumably just go under.
Horses absolutely are consumers. They eat, they drink. They just can't manipulate their environment to get more food or drink.
The human system relies on us doing stuff for each other, mediated by "money". We get "paid" by some other human for doing stuff for them. "Stealing" is discouraged. "Being given" is rare. "Money" originates from fiat, goes where it goes.
Humans unable to do for others don't get "paid", get starved/sick etc, experience pain, cause trouble and disappear eventually. When doing stuff for people is not done by people, the pay cycle breaks. When machines do a substantial fraction of economic activity (intangibles like decision-making, writing, answering the phone, etc. as well as transport), there is no need for very many people. Absent the need, there will be no support. The farms won't give their food away, they will arm themselves against the hungry.
There will be some human work. So long as machines cannot equal human touch and experience combined, there will be medicine. There will always be sex work. But who affords it?
Possibly racial disparities will diminish. A machine might actually give equal medical care to, or equally arrest, the Black and white person. Or not, depends on the training. Look who's developing the AIs.
Economists thought that economic development means democratic development. They were wrong, see China. They thought that economic liberalization means good governance. Wrong, see Russia. Economists are right about things until they are wrong, and we can all run off the cliff with them.
When the population drops (war, starvation, refusal to bear children) to the level needed to run the machines, the system will stabilize. Pretty much the horse scenario.
Yes, we've seen that already without necessarily a technological singularity. You accurately describe the great depression in the 1920s and several similar events prior to that.
But what's consistent with those events is that they end and economic development is resumed under different circumstances. I don't see any reason that the current era will be different or that current problem will be unfixable. People used to get paid to build national parks. What would be our current iteration of national parks?
This transition will be a lot slower than the shocks that produced depressions and recessions. A couple of good wars will thin the herd, which is how populations have been weeded out in earlier times.
Public works projects need funding and purpose. I'm not optimistic about what purposes the make-work funding gets applied to these days. Public beauty, cleanliness, health are not in the priorities.
Huh? The Great Depression was a cyclical event. So public works projects only needed to last until the market recovered. Technological revolutions are structural changes. There is no “national parks” equivalent program which we were willing to run to sustain the horse population at 1900 levels.
Does this result in millions of micro-businesses? All on the verge on bankruptcy, all continuously going out of business as new ones are invested in by VCs hoping to make a quick buck
More likely it results in a state of permanently doing gig jobs as any temporary work you can find is just a front to train a lower quality, lower cost automation to replace yourself.
That is one possibility but it goes against the interest of all income groups of the economy, including the ones on the very top. So a solution is in everyone's interest to be found, and nor is it a physical impossibility.
Until the ones at the top own a fully automated supply chain for their every need and luxury and all they need from everyone else is the ability to feel superior.
In this case, why doesn't the rungs of the economy below the rich just... run the way they have been? Serious question, I've been mulling over in my head what it means for a few people to own the majority of "the economy" and goods to be very cheap, but no one has any income.
What I always come back to is this: Why does it matter what the very richest do? Worst case the economy operates approximately as it does now (maybe with us knowledge workers knocked down a few rungs). But like, the rich can't eat _all_ the food.
I feel like the most likely outcome is that the quality of life of the bottom 90% is overall better than today. HN'ers quality of life is maybe worse than it is today.
They keep buying ten of the houses I want to live in, knocking them down, and building a single tasteless McMansion on the combined block, and then lobbying for building and zoning regulations that won't let me build another house.
Presumably if we're all unemployed or employed in jobs that have zero economic worth to them, they'll buy all of our houses, build a bunch of big grey boxes for us to live in like a human zoo, build even bigger and more tasteless McMansions for themselves, and pass legislation that makes it illegal to live near them or inconvenience them.
The quality of life for the bottom 90% of the world has been getting better because they were so desperately poor that they could rely on economic growth to help them. At some point they're going to be rich enough in commodities that they start looking to buy things like houses or start businesses, at which point they'll find that asset prices are rising faster than their own wealth because of growing inequality, and they will look at their big grey boxes full of shoebox apartments and start to become unhappy too.
I think if we jump instantly to a fully automated world, perhaps there's no problem, but in a normal transition from the actual world the issues are
- capital / limited goods
- allocation of labor
Things like land are a limited commodity. Concentration of ownership usually comes with rent extraction. Because you need to live somewhere, but available locations are finite (and supply can even be controlled, and is). This means that you can end up with a cartel scenario where increasingly normal peoples previously disposable income is tied up in an increasingly large % to those fixed costs of living. This is arguably well under way in some countries.
The second point is, the richest enjoy goods like personal yachts, private planes, luxury restaurants. If a lot of labor is increasingly allocated to producing highly inefficient luxuries, it's essentially wasted. You might have hundreds of businesses employing thousands of people or machines, but the productivity of their work is essentially all going into products consumed by a few people. I'd say a limited amount of this is a good thing, it's a normal part of the proliferation of choice enabled by capitalism. But taken too far you return to a historic economy where most peoples productive life is spent serving a tiny elite, while living in poor conditions themselves.
In our financial system, your income as an individual has to be significant enough to send a signal to produce the goods you want. If this ceases to be true for even a significant minority of people, you will find consumer choices and available quality decreasing for those people, which is a real problem. This is to a lesser extent happening in some segments of the economy.
Yes, though I'd say that rent extraction and labor being directed to the whims of a small number of people is essentially the same thing. And even under a step jump to full automation, the owners of fixed goods can still extract monopoly rents.
This is also largely the conclusion of classical economics (i.e. Adam Smith) and the argument was developed in great detail by Riccardo, George (who explicitly considered the no-labor-required case in 1800s), and many others.
Narcissism and psychopathy are why. The richest people stopped needing to care about money to fill their lives with endless luxuries billions of dollars ago. They enjoy being richer and more powerful than others.
The rich will buy all the scarce resources and charge the maximum possible price to extract all of the value you can provide. You will be unable to accumulate wealth. Dynamic pricing of fast food will extend to all food and even to e.g. plumbing fittings - you don't own a copper mine, so you can't go make your own. You will do hard and degrading work for a subsistence living despite the ample resources and technology required to improve your condition.
This is history, not a random hypothesis. Feudal serfs had the ways and means to bootstrap, but in fact lived worse than their pre-feudal ancestors.
I actually have a random thought that it is this innate human drive that causes society to be mostly employed despite massive differences in technology and productivity across geography and time.
What if the driver of employment is not efficient return of capital, but satisfaction of human mind?
Your statement makes it seem like we had no problem losing professions to automation. I'd argue this isn't true.
There are plenty of poeple who have inadequate work the world over because there isn't enough paid work to do, and most professions are servicing wants, not needs, so the demand is very elastic, with many people now multi talented and still not making ends meet.
I know it's been talked to death already, but the fact that the video is so wrong on self-driving cars is mind-blowing. We can rationalize post-fact as much as we want - still it is shocking that popular belief never expected language models before self driving.
Popular belief never took the regulatory obstacles to self-driving cars seriously. If LLMs required insurance to use, they also wouldn't be available in public yet.
I'm not sure I agree that it was ludicrously wrong. His thesis is "the question isn't whether self-driving cars will replace human drivers, but how quickly". Sure it's been slow going, but you can't walk down a street in San Francisco today without seeing a Waymo, which wasn't true a few years ago. We're still at the bottom of the S curve.
1. There are no rules of economics. Economists attempt to understand things that happen and use that understanding to predict future events, but frankly there’s not a lot of consensus.
2. Economics is a function of human behavior, not horse behavior. Economics is about coordinating human effort. Humans make economics, horses don’t. When economics change in a way that is bad for horses, the horses have no recourse. When economics change in a way that is bad for humans, the affected humans seek to change things.
Indeed. The economy may in fact eventually close in on itself and become a fully automated system with no humans in a loop. "Disneyland with no children", as I believe Bostrom called it.
Assuming we are around for this, this raises the question: if the economy can function without us, what claim will we have on its outputs?
Isn’t technology defined by how things can be used by humans, I.e. tools? Horses can’t use smartphones, and a “smartphone” designed for a horse would look significantly different. Seems a like a nonsensical argument based on the definition of words. Technology is what humans can use to achieve tasks.
They still are, wherever people work with animals at small scale. People get attached, and the boundary between "animal as my fellow (if dumber) friend" and "animal as my main capital investment" gets really blurry. This was the reality for most people until 20th century, and we still see echoes of that in old books and stories.
This sounds like a clever argument, but the mindset of some executives really has no impact on how differently horses and humans are treated in society.
To spell it out:
When horses are no longer needed for a task, they are sold or sent to slaughter.
I hope I don't have to explain how differently you and I are treated when we lose a job?
People need jobs in order to pay for food, shelter, and healthcare.
Many executives feel zero moral obligation to provides jobs to people and instead treat that as a necessary evil that gets in the way of profit. As soon as that evil becomes unnecessary, they will happily lay off as many people as they can.
In aggregate, when enough executives do that, then people will not be able to find jobs.
When they can't find jobs, they are unable to pay for food, shelter, and healthcare.
I don't disagree. Trajectory has overall been great for people. Just saying it's not uniformly great for everyone at all times, in fact whenn it regularly flies off the rails, restoring it exacts a great cost from individuals and societies.
It's not really "many" centuries. It's more like three or four and even that's not uniformly distributed. There are still many subsistence farmers in the world, and almost all of us were doing that for thousands of years.
Also, those past few centuries have involved the largest violent revolutions and wars in human history.
The idea of employment and money is a relatively modern conception. There is no physical law that states it must work indefinitely and efficiently for all participants.
If software engineering goes the way of the dodo, I suspect most other white collar jobs are also displaced (anything done entirely on the computer, really). A few years after that I suspect an overwhelming supply of physical labor.
It’s a house of cards at that point, sure - software engineering could be the first to go, but I don’t see any reason to try to retrain or pivot. As soon as I lose my job and jump to plumbing, the rest of the white collar jobs are going with me.
If we get to that point, I’m just going to assume that the rest of the world is also going to be in shambles.
Though, I’m more of a Jevon’s paradox believer. We’ve had so much work in the backlog and now we can finally take a stab at it, I’ve been busier than I ever have been. Until the AI can competently run our entire stack, I’m skeptical. If it gets there, I’ve got no reason to try swapping jobs, because everything will be busted in a short timeframe - no?
I’m also trying to keep my eyes on the big dogs. Once Google starts to hemorrhage engineers and knowledge workers that means that AI can meaningfully replace entire roles, but until then I’m just sitting tight and building stuff, business as usual I guess…
I feel for everyone else in this position, on edge for the last 3 years with no end in sight.
> If software engineering goes the way of the dodo, I suspect most other white collar jobs are also displaced (anything done entirely on the computer, really).
Strongly disagree with this. In contrast to most jobs, software engineering is unusually self-contained, with short feedback loops and immediate validation. You (or Claude) implements the feature, you run the tests, you fix the bugs.
Most white collar jobs are nothing like this: lawyers have to prepare briefs (AI can help here) but they have to file briefs, they have to wait weeks or months for rulings, they have to argue in court. Marketing execs might spend millions preparing a new campaign, launch it for the holiday season, and it has to be right the first time. Most white collar work involves judgement, anticipation, and ambiguity in a way that software doesn't.
Software lets you guess-and-check over and over again until it works.
> Most white collar work involves judgement, anticipation, and ambiguity in a way that software doesn't.
In my current position as a software engineer, this is MOST of the job.
I don't know what everyone else's role is like, but a software engineer that isn't dealing with those three items sounds more like a compiler.
"Most white collar jobs are nothing like this: [software engineers] have to [implement features] (AI can help here) but they have to [coordinate releases across teams], they have to wait weeks or months for [real-world usage to validate architectural decisions], they have to [defend technical decisions to stakeholders]. [Engineering teams] might spend millions preparing a new [platform migration], launch it for the holiday season, and it has to be right the first time. Most [software engineering] involves judgement, anticipation, and ambiguity in a way that [isolated code production] doesn't."
Sure, and to the extent that your job involves those things, it won't be taken over by AI as quickly. But this whole thread is full of software people screaming that the sky is falling and Claude can already do their entire job and they are the horses, not the farrier. So...which is it?
I think software development is a field unusually vulnerable to LLMs because there's lots of good examples for training data, and the work is easy to validate in short order according to somewhat-objective criteria.
Note that certain kinds of mathematical proofs also fit these criteria, and also have been an early showcase for AI abilities. Anything where the bot can get to a good result by just grinding on it for days is going to get taken over by AI quickly.
> You (or Claude) implements the feature, you run the tests, you fix the bugs.
...you then take it to the customer. And wait, and wait, and wait, and maybe eventually they'll get around to trying out what you created, and if you're lucky they'll even give you some feedback to loop back with to try again. They will only try your application a few times before giving up on you, so even if the feedback loop is immediate like you claim (its not; not even close), you aren't afforded many iterations. A software engineer's job is, of course, to get ahead of the customer using judgment, anticipation, and dealing ambiguity to minimize how many round trips are required to satisfy the customer.
If AI can replace the software engineer, why can't it replace all "white collar" work?
For the record, I don't think AI is likely to replace those parts of the work, at least not in the near term (15-20 years). But at many companies, the people interfacing with stakeholders or users — deciding what problems need solving and how to solve them — are product owners or project managers, not software engineers.
I think there will be a convergence where AI enables more POs to create software directly, and more SEs move up the value chain to work on product-level problem solving.
Product owners/managers typically act as the communication channel between customers and software engineers, but engineers are responsible for designing the software. It is they who have to use judgment, anticipation, and deal with ambiguity in order to find software customers want to use. The product owner/manager does not. They simply pass messages back and forth. Of course it is possible for one person to be both a product owner and a software engineer, but often the engineer is separated from having to play telephone tag with customers so that they can more efficiently focus on their primary task.
Perhaps you have confused software engineer with coder? LLMs are most definitely the final death knell for anyone still working as a coder, but that profession was on the way out for decades already. The earlier comment was clear about software engineer.
Yeah, most PO/PMs I've encountered know awfully little about the business processes, their constraints and complex interactions within their own product. They know quite a bit about priorities the various customers voiced, fancy stuff they'd like to demo on the next review or trade fair or customer relationship meeting, or new features they want to get customers interested in. But don't ask them how those features interact with others!
If you want to go into detail with that kind of stuff, your best bet is not the PO, but the lead engineer of the team. If the team was properly staffed, the lead engineer will have a solid knowledge of the business aspects of the product while at the same time being able to cross-reference that with the most important technical aspects.
LLMs will make more people into "lead engineers", even if all they're "leading" is the LLM agents and not other humans, as it was common before the LLM age. One of those will then emerge as the "lead lead engineer" who's going to lead the humans which in turn lead the LLMs. If it wasn't for the "engineer" in the meaning of "code monkey" dying out - which I also expect - we'd have to invent a new title, but I'm currently expecting this shift to go without that. It'll still leave some engineers who made themselves comfortable just churning out code that does whatever a ticket requests, without even understanding one iota of the business process behind the ticket, without a job, but that's the kind of "software engineer" who in my opinion shouldn't ever have taken up the profession anyway.
> Of course it is possible for one person to be both a product owner and a software engineer
This was the norm in every job I had from late 90s to around 2015.
Engineering Managers were responsible for both technical and business outcomes, and it worked fine.
The scrum solution of having from the business side acting as a literal "product owner" or "client representative" and giving feedback IMO was a much better direction that unfortunately companies have not followed.
I don't think "software engineering" is a real thing. So-called software "engineers" do not practice engineering in the way that a mechanical engineer or a civil engineer does. Consider: How often do bridges collapse? How often do elevators plunge to their doom? How often does some critical piece of software utterly fail? [0]
> They simply pass messages back and forth.
So you claim. We're about to conduct a real-life experiment to see which is more valuable: passing messages back and forth to the computer, or passing messages back and forth to the customer.
> I don't think "software engineering" is a real thing.
You don't think anyone designs software? It just magically appears?
> So-called software "engineers" do not practice engineering in the way that a mechanical engineer or a civil engineer does.
And? The dictionary defines engineer as someone who designs and builds things. It does not prescribe how the thing is designed.
> How often do bridges collapse?
Not often, but often enough to know that we still consider the designer to have been an engineer and not a project manager. I still have no idea where the idea that project managers do the design work came from.
> passing messages back and forth to the computer, or passing messages back and forth to the customer.
When AI can handle the judgment, anticipation, and deal with ambiguity neither are important. The customer will go straight to he AI. Hence why "white collar" jobs of all sorts are ultimately doomed. Unless you believe AI cannot reach that level, in which case software engineers will be fine, since that's the whole reason for their existence.
What we do see happening already is the decline of the project owner/manager role. More expensive money and technology increasing engineering efficiency has made it such that it is becoming hard to justify an assistant to bridge the telephone tag gap. That doesn't really affect engineers much, though. Before ZIRP allowed budgets for project managers/owners engineers also played that same role, so it is nothing unusual for them. Engineering getting secretarial assistants for a time, something historically reserved only for the upper brass, is but a curious aberration.
> You don't think anyone designs software? It just magically appears?
Do you actually write software for money?
Most commercial software is never designed, it evolves, over the course of repeated "agile" "sprints". Every two weeks, someone hacks together something that kinda-sorta works and then that is released.
Eventually (often many years later) the thing becomes so complex and unwieldy that it's impossible to add any more features, so someone gets angry, throws it away, and starts again growing a new monster.
No. I am paid to be involved in the software industry, but as an engineer, not as a coder. Writing software is not part of what I do. That is, to use your bridge analogy, for the construction crew.
> over the course of repeated "agile" "sprints".
It is true that not all engineers know what they are doing, are good at what they do, or bold enough to push back against out-to-lunch stakeholders.
> so someone gets angry, throws it away, and starts again growing a new monster.
I never understood that. Untangling the messes was always the fun part back when coding was still a job. But I don't really see any coders in industry anymore. Any last remaining ones I knew of have already been replaced by LLMs. LLMs will also happily untangle the messes, though, so still no reason to abandon code.
Boeing ships software that crashes planes, Microsoft Windows regresses so you can't move the start menu, medical devices come with hardcoded admin credentials.
>> One quirk of the program was that the amount field came pre-populated with 15 zeros, which the person inputting a transaction needed to delete, something that did not happen.
> One quirk? Fifteen zeros? The quirk in the software is that if you type “send $1” it says “okay sending one quadrillion dollars” and you have to backspace over all the zeros? Incredible interface design.
> I'm very very curious where all these mythical software engineers are working.
Boeing, Microsoft, and medical device manufacturers mentioned seem like pretty good candidates. Were you under the impression that the software designed at those places just magically appears?
As before, being an engineer does not imply being good at the job. It only implies designing things. The designation of Professional Engineer institutes a vetting process to try and weed out those who aren't good at the job, which may be the source of your confusion, but PE is a designation, not a job. Not all PEs even work as engineers, although certainly many do.
Ok so why is all their software such garbage, then? You can't gaslight me, I work in this industry. I see the absolute nonsense that passes for marketable product. I know there are smart people involved in creating software, but they are not engineers and (with the possible exception of some groups at NASA[0]) nothing they produce is the result of an engineering process.
Has any software "engineer" ever lost his license because he signed off on trash code that failed? Where is the Otis elevator emergency brake[0] of the software world?
> Ok so why is all their software such garbage, then?
Because the people involved weren't good at their job...? We've been over this multiple times now. It doesn't seem like it should be hard to figure out. What do you continue to miss?
> but they are not engineers
Sure, software involves lots of different types of work. Engineers are just one tiny piece of the puzzle. But design is exclusively the domain of engineers. Anyone working on the design is an engineer, by definition. That, again, does not imply that they know what they are doing, are good at their job, or bold enough to push back against out-to-lunch stakeholders. Having someone doing engineering work does not magically mean you are going to see an excellent product on the other side. Most people are pretty lousy at what they do.
You, again, increase your chances if you involve PEs, but that's not a job, just a designation.
> Anyone working on the design is an engineer, by definition.
You and I have very different definitions of "engineer". To me, an engineer is a licensed professional who adheres to the code of practice for his or her profession. We'll have to agree to disagree on the semantics. I think lots of software people want to call themselves "engineers" because it sounds like someone who should be paid more money.
Again, we’re talking about jobs (and whether they will remain in the future), not professional designations. Licensed professional engineers are usually referred to as PEs, but even if you want to call them engineers it is clear that not all engineers of that sort work in engineering. In context, we can establish that we are talking about engineers in the worker sense, not the professional license sense.
Is English a second language for you? Maybe context doesn’t translate well. English is certainly bad for assigning multiple meanings to words and relying on context to figure out which meaning applies.
> I think lots of software people want to call themselves "engineers" because it sounds like someone who should be paid more money.
How would describing the job by what it is, literally by dictionary definition, garner more money?
I am aware of the time physicians started calling themselves doctors to try and ride on the coattails of academics, so I think I understand the angle you are trying to come up with, but this isn’t like that. In this case engineer was already the word in the dictionary that existed to describe the work.
Even outside of "software engineering" there are tons of people working jobs that almost everyone would consider "engineering" that don't require a stamp. Chemical engineers, electrical engineers, mechanical engineers, tons. Only ~5% of chemical engineers have passed the PE, I guess there are essentially no real chemical engineers then?
Even in a civil engineering firm, most likely the kind of engineering you're talking about, only about half will actually be licensed and have a stamp. Tons of people will work on a project and figure out a lot of details, then work with the person with the stamp to sign off on the implementation details. Their work is undoubtedly still "engineering", but they don't have a stamp.
Engineering is ultimately all about managing tradeoffs, so the more important question is: How often does reliably picking failure modes and applying appropriate safety factors actually matter in software, especially when most designs are little more than simple line-of-business CRUD with a (if you are lucky) slick UI?
>In contrast to most jobs, software engineering is unusually self-contained, with short feedback loops and immediate validation.
This is nice and makes things easier for the models to learn. Emphasis on easier. They are still capable of learning how to deal with other kinds of problems and being better at it, and they have, over the last couple years.
I'm a little surprised by how easy people joke about residential electricians and plumbers in this context. To begin with it requires young and healthy back and knees. How many men after 40 and especially 50 are like that and for how much longer?
Secondly, in CA to be a real residential electrician you need to be C-10 licensed. Which requires all kinds of time consuming prerequisites to begin with. Some of them could be a challenge if you are not the kind of guy who knows a guy. Apprenticeships, unions, depending on your major 1 to 4 year of "hands-on, full-time journey-level experience" and such. Your best hope is probably that your EE degree from 20+ years ago still counts for something.
And there quite a few 20 yo candidates to compete with who are taking DC/AC circuits classes in community colleges this very moment. They are not spoiled with life-long office work experience.
The competition among people who'd like to be an industrial robotics technician at the Tesla Fremont Factory is very real too.
This is generally how I feel - whenever new levels of abstraction have been introduced the goal posts for what can be built are now orders of magnitude further. This seems to be the case with all technology.
Consider assembly -> higher order languages -> libraries -> open source -> stack overflow -> now AI.
> If software engineering goes the way of the dodo
Much more likely is that software engineering goes the way of the elevator operator. Advancements in technology saw the number of elevator operators explode, to the point that we've all become one!
Unfortunately for those wishing to make a living as an elevator operator, due to the properties of supply and demand, us all joining the ranks of elevator operator means that supply is effectively unlimited and thus price cratered to effectively zero. This is the real risk to software engineers, at least those who are in it for the money.
I guess the question is more if you're a coder or a builder.
If software engineering was about writing syntax and flipping bits for you, this might no longer be an economically viable activity.
If it was about building useful or fun things, then AI will probably not take that away, even if it's just for the simple reason that you care more about it than some hot GPU.
Yup I agree. I do think that all the low(er) skilled software jobs will disappear. Those mostly working on marketing websites or simpler CRUD apps. And if they don't disappear, they will just not need nearly as many people for those anymore. But those jobs were mostly about writing code, i.e translating a Jira ticket to reality, and that's not really "engineering" as such, so if your only marketable skill was knowing how to turn English (Jira) into Syntax (Code), then yeah, better start upskilling fast.
The engineering part of software was already completely dead for a large portion of coders who didn't want to know anything about the domain, left all the decision-making to business people/product managers/designers and was downright allergic to doing anything that was more complicated than a CRUD in their favorite framework (which would invariably either change daily or remain static for their entirely career).
There was a huge influx of people like that in the last 10 years who got into it for the money, and frankly the industry got much worse and much more hostile to different perspectives and different experiences.
People who have not evolved into builders have visibly become meat interfaces to the LLM.
Have any of these articles actually addressed HOW software developers are supposed to retrain for good jobs? Personally I have neither the money nor the years to go back to college to get a livable job.
I know some people whose work was already almost completely eliminated by AI (language translation) and they're not doing great. Reality is there's no-one out there looking out for you except yourself and very close family/friends.
Oh yes, when LLMs fail at translating they don't make a visible error. They fill up with inferred meaning. Translators can only be replaced by LLMs if there's no consequence attached to the meaning of what's being translated.
It could. But I'm not confident there will be a broad recognition of the fact if things become much worse than they are, or of the cause, or that there will be a desire to make things be better. There has been a long decline in quality in favor of optimizing for monetization and engagement, predating LLMs. But I do think there will be some pullback in things that actually matter if not in the SaaS FOTM.
kinda. there'd still be a lot less demand for software devs than there used to be, a lot of people would still be jobless/would have to pivot to some other career.
So obviously each college is different. And your experience is not mine.
But my college did not train me to get a job the first time. Indeed they explicitly explained to us that that was not their goal. That was helpful because up to then we did think they were doing job training.
What they actually set out to do was make us "smarter" people. In the sense that they taught us to analyze problems, work in teams, find creative solutions. They taught us about research and critical thinking, and logic and so on. How to hypothesize, test, evaluate, adapt.
You don't need to go back to college to learn all that again. You already know it. Skills are easily, and quickly taught and learned. That's the easy part. The hard part is approaching a job as a problem to solve- how and where to add value. Value people will pay for.
Look around you. See unmet needs. Figure out what you need to do to meet those needs. That's the real challenge we face going into this next era.
I don't know where you are, but when I hire mechanical engineers, I hire them from universities with accredited engineering programs. It's actually a legal requirement for the job in my jurisdiction. And if those engineers decide they want to become daycare teachers or knee surgeons, they do have to go back to university and re-train, even though they already learned how to think the first time.
You're right. There are several occupations which have legal entry requirements. You can't be a doctor without getting a medical degree.
However the number of professions that require a formal degree is very small. (And notably, programming is not one of them.) If you take any large group of random people and asked them, you'd find close to all are not required to have a formal certification.
Generally yes, people with (any) degree are more hirable to (any) job. The skills getting a degree are generally useful, and make a candidate more employable. But if you already have one, you already have that.
Is that true for healthcare careers that don't have abysmal quality of life? All of the nurses I've known are working 60 hour weeks, night shifts, etc.
If you're unemployed, you won't need to pay 2 grand/month for someone else to watch your children, and you don't have to pay San Francisco rents if you're not working there.
I suppose you've got a point; if the middle class of Americans constantly have their careers destroyed, eventually we'll return to a third world sort of existence where childcare is provided by extended relatives.
However, most people in America can't rely on grandparents for childcare. One reason is we have children much older than people in the third world do - if two generations have children at age 38, it's not very reasonable to be expecting a 76 year old to provide child care.
And I suspect disaster to have correlated chances of killing parents or grandparents or children. i.e. if you've become a parent there's a good chance your parents did too.
Our own lived experiences are likely outliers, since developed countries have different outcomes than the past.
I don't think "just give up your existing community and move" is a viable option. That defeats the whole point since at that point the retraining isn't allowing you to maintain your quality of life.
It is a viable option. "Quality of life" is a luxury and programmers no longer belong to the class that can afford it. Do what everyone else does and go wherever the jobs are and live at whatever quality the market decides that you deserve.
I think you are missing the point of the discussion. The idea behind "just retrain" rhetoric is effectively "new technology won't destroy people's quality of life because they can just retrain!". If quality of life isn't maintained, then there is no reason to retrain at all, people can just get a minimum wage job and live in poverty.
The rhetoric was a lie. Capitalists lie. That new technology always creates a surplus of newer, better jobs and quality of life always improves for everyone was a lie. That you can "just retrain" and everything will remain the same was a lie.
Even if you "retrain" your quality of life will go down because the market value of your life will go down as automation and AI improve, unless you're a member of the capitalist class. Which, if you need to work for a living at all, you aren't. No matter what you do unless you're personally wealthy your quality of life will go down. That's just what an efficient market does to most people, and AI is going to make the market brutally efficient.
I wonder how many sectors this will have to happen to until workers realize that nobody's safe. I was shocked to see journalists at major firms peddling the 'just learn to code' line a few years back, given how much their industry had contracted in the decades before.
Quality of life during the Industrial Revolution dropped considerably for most of the working population, but afterwards saw big improvements. You seem to be implying that life now is worse than before the IR. Are you? If not, in what way was the loss of jobs to mechanisation then more acceptable than it is now?
The overall quality of life improved through and thanks to IR. But the quality of life of the people most affected, and their children, and their children's children - did not do so well.
Technological progress automating away jobs lifts humanity as a whole, but it absolutely fucks over the people whose jobs got automated away, and the families of those people, and the communities they live in.
San Francisco is where the jobs are, and dropping your entire social network on the floor at exactly the moment you need it the most to help you is not good advice.
Former tech workers can ask for advice from non-tech people who previously had to leave because they could no longer afford rising costs that happened due to influx of well paid IT jobs.
No, it isn't. This is an incredibly modern phenomenon. We see examples of peoples moving en masse throughout history, yes, but the idea that everyday individuals should expect to leave their community is an incredibly recent one. Such would have seemed monstrous to people even a few hundred years ago. Fun fact: Socrates wasn't actually sentenced to death, he was sentenced to death or exile, and chose death over having to leave his home behind.
The entire colonization of the planet by humans from 250,000BC to today would argue otherwise
We’re physically built to be mobile - the extent you can take your community with you has always been relevant
I’m sitting in Virginia USA with 50% of my neighbors being migrant families of 5+ people usually across three generations
I’m not sure what data you’re working with. In fact there a whole lot of ignorant politicians that want to stop migration precisely because of how valuable of a tool it is.
Some humans moved their families from somewhere that they weren't safe, to somewhere that they could be safe, with the intention of settling there and building a living space their great-grandkids could be proud of.
Some humans lived nomadically, moving all the time, as a way of life.
Being forced to do the second because nowhere is stable enough to do the first is what we're talking about.
Migrating is one thing, having to migrate another.
no one leaves home until home is a sweaty voice in your ear
saying-
leave,
run away from me now
i dont know what i’ve become
but i know that anywhere
is safer than here
I'm assuming you're talking about people born in SF? Because the rest have already demonstrated a willingness to give up their community once in the pursuit of quality of life.
Equally, if you are in SF you're about to feel the pain you've caused to the locals by moving there. Somewhat harshly, I could suggest that you didn't care about their existing community when you priced them out of their neighborhoods.
So here's the rub. SF is somewhat different. It is a company town. If you don't work for the company you're moving. Sorry. There's certainly nothing else that can pay you the same money.
On the up side, there are lots send lots of cheaper places to live. And chances are you can find a good job there doing something that adds value to that community.
Only $2.5K rent? That's like free these days. For what though. I rented a 4K sq ft house in Charlotte three years ago and now paying $3.5K for not a modern 2bdr in Miami proper.
Because that's how they are built in the US. I didn't need the extra space, but for the money, there wasn't much choice, really. I would have loved to rent a house half smaller at half price, but none were on the market. I barely went upstairs. My dog occupied it, at least.
I could have rented a townhouse for $300-400 less, but then there are different constraints. Small or no yard, tiny kitchens, no garage etc.
I was perfectly content with a 2200 sq ft ranch in Henderson Nevada.
That's just sad. I knew that real estate was a pain in most countries around the world, but I never realized that choice in the land of freedom was that constrained. I guess in most places it's basically: "Detached House, Large".
The Charlotte house was (2022ish) $2.3K/month through a PE that bought out a lot of single-home houses to rent out. 2 story, 4bdr, 3 baths, a yard and a lawn, a huge deck. The area was very nice and so were the neighbors. My dog loved the deer.
It wasn't urban enough for me, for someone working remote at the time, and few growth opportunities for my wife, but I regret moving to FL. Miami has become the least affordable city in the US relative to income.
I keep telling my 22y who graduated from a university last year to go to a trade school (HVAC/ electric, plumbing). Too late for me as these require apprenticeships in the US.
My HVAC contact in FL needs a $100 just to sniff the AC/boiler closet. That is, if I can get him to show up. I don't even know if he's licensed here, but he's definitely very experienced.
This works if you're interested in running a business, but what well paying work doesn't require a degree or some kind of formal training certification?
First you assume that being "interested" in the work is a requirement. I'm guessing 90% of people are doing work that doesn't interest them. But assuming that it's a requirement for you the question becomes, what are you interested in?
Second you want well-paying work. I understand. Programmers got well paid. They lived above the standard of most. It's not surprising you'd want to keep that. But "well paying" is a spectrum. It depends on your life-style, location, eating habits and so on.
And finally, lots and lots of jobs don't require a formal certification. Or one that isn't trivial to get. Ask the people you see everyday (including the software programmers, hardware designers, receptionists, assistants, book keepers, marketing and advertising people, sales, support, your vehicle mechanic and so on.) I suspect if you asked every dingle person you met in a day the ones that actually had this requirement would be a rounding error.
I know what you’re trying to say, but strictly speaking there are many examples of “hard-science” people moving towards and excelling in softer fields, while the opposite is almost never true.
There was a saying where I went to high school: “the language programme students are not better at languages, they’re just worse at math”.
The language students were however better at writing and communications in general. They ended up in sales, marketing, advertising and everywhere else.(all under threat from AI as well.)
But to further the GP question there are lots of examples of people not working in their degree in both directions. The entire computer industry was built by people without a Computer Science degree. Today huge numbers of programmers have no degree at all.
To refer back to the original post about the farrier turned mechanic. Most comments seem to miss the point. Change is coming. You can lean into the change and prosper. Or fight the change and vanish. Those with enough vision, creativity, adaptability will continue to thrive.
The same way clerical and manufacturing workers retrained when their jobs were replaced by computers/the Internet and robots. That is to say, most won’t.
People in tech change roles and careers all the time. The vast majority of role or career changes do not involve going back to college.
I would even recommend against going back to college unless someone has an extremely compelling case for a career change that requires it and the conviction to see it through.
I'm hesitant to give examples because I know every example I give will be met with objections from "But I don't want to do that job" or "What if that gets replaced by AI too?". Look around you and you can should be able to identify some adjacent roles where you at least have an understanding of what the works like and you could learn the rest by taking on gradually more responsibility.
There are a lot of jobs out there that aren't exciting, don't pay SWE level salaries, but nevertheless pay enough to support a family as evidenced by all of the people doing those jobs who support their families. Not every company or industry is interested in squeezing maximum efficiency out of their employees. I have friends in highly bureaucratic roles who care about their jobs and do it well, but their work is amazing simple and easy. It requires nothing more than being a responsible adult who can show up to the office and get things done. These jobs are everywhere and I don't think they're going to disappear.
I don't think SWE is going to disappear either, but it will look a lot different in years to come. If you can't or won't adapt to the new normal, one of those boring alternative careers might be worth pursuing.
>The vast majority of role or career changes do not involve going back to college.
That was possible pre-2022 during the famous developer labor shortage. In the current market, companies expect seniority/experience in that specific stack they use for that, no more career changers.
They either take you as a senior/expert to contribute from day one, or as a young and cheap newgrad/junior out of uni they expect to train on the job, but no more in-betweeners switching careers as they're not as experienced as the seniors on the market, nor are they as cheap and malleable as newgrads/juniors.
Sure, in theory every developer can learn at home to do anything new related, but in the current job market, companies expect X-years of proven on the job experience with their stack, no more bootcampers, and no more self-taught-at-home types. The entry bar has risen significantly. At least where I live.
Like I just saw on linkeidn a job posting from a large bank that was an hour old and it already had 100 applications, which probably means most applications are now AI botted, meaning yours will get lost in the noise and getting a call back for an actual interview is like rolling a dice.
Not saying it's impossible to change(I did a couple of times), just that's it's much more difficult than in the good times, and will most likely also come with a paycut.
To make matters even worse, if you do apply to these jobs the only thing you get back in return is your data being sold to even scummier organizations. I always use single-purpose emails for everything, so I know where things source from. Imagine my shock when a week or two after applying for jobs I started getting a steady stream of Super Intelligence emails offering to spruce up my resume, or to source leads for me, or some other sketchy service.
Everything is commercialized now. Even the discard pile of resumes can be sold for pennies. A job application isn't a job application anymore: it's a desperation signal alerting predators to come extract some value.
I guess where I'm coming from is that I currently am trying to change careers. Everything I'm finding is either a 50% pay cut or is simply unwilling to consider anyone without 5-10 years of experience in that particular industry. Since I've already built a life here I can't really just move and a 50% pay cut isn't enough to live anywhere near a comfortable life; particularly since I have medical expenses that aren't covered by insurance.
> I have friends in highly bureaucratic roles who care about their jobs and do it well, but their work is amazing simple and easy. It requires nothing more than being a responsible adult who can show up to the office and get things done. These jobs are everywhere and I don't think they're going to disappear.
I think these jobs are actually the most likely to be replaced by AI. They have resisted automation to date because there's too much fuzz in the inputs/outputs to make it feasible to computerize. LLMs solve exactly that problem.
Except for the part where they have absolutely no idea what the world is or how it works. They are good at software and math because of RLVR, but they can’t run a business.
>They are good at software and math because of RLVR
Software and Math are not the only things LLMs are fast improving on.
Astra is one of the biggest examples of this being untrue. The model is cracked at computer use and vision/spatial tasks and reasoning. Driving, Operating Robots, Video Editing, 3D modelling, graphics are all things Astra was >>> at than any other model.
>but they can’t run a business.
And it's the leader on VendingBench. Also most people cannot run a business.
Kinda true (they have a mediocre idea of how the world works), but it's surprising how close they are given their general limitations and what it took to get disruptively good at software and maths.
Like, sure, using an LLM as a CEO today will bankrupt you and some of the decisions leading to that are daft, but even despite that it's in the same "huh, interesting" kind of zone that early LLMs like InstructGPT and the first release of ChatGPT were with code. It could write short functions or small scripts for you, even then… but any attempt to go beyond that would fall over quite fast.
Astra told me I can walk to the car wash to wash my car.
Again, the improvement in coding is due to RLVR, and it does not generalize to other domains. So it is meaningless to point to examples of that improvement.
There are many ordinary behaviors and skills that are not feasible to RL at scale. The sampling efficiency is just way to low.
> Astra told me I can walk to the car wash to wash my car.
Yes, like I said, a mediocre idea of how the world works.
> Again, the improvement in coding is due to RLVR, and it does not generalize to other domains. So it is meaningless to point to examples of that improvement.
I can point at actual benchmarks for performance on business tasks. On the one hand, those lines go up; on the other, I just don't know how good those benchmarks are.
> There are many ordinary behaviors and skills that are not feasible to RL at scale. The sampling efficiency is just way to low.
Yes. A point I make myself on occasion. One of two perfectly valid ways to describe intelligence, and one by which all current machine learning really sucks.
Like, if it was fully general, the trillion or so miles that Tesla cars have driven would have obsoleted the steering wheel long ago, rather than the closest we have being (yet another) pronouncement from Musk that this new car he's not actually made available for sale won't need one.
Business self-play training could be anywhere from code to cars. I was expecting much faster progress with the cars, and much slower progress with code, so I don't know how to adjust my expectations with any other domain, including business.
It’s much harder than both code and cars because there is both more variety to the work and much work where the reward function takes months or years to resolve objectively.
In general it could well be as you say, but the current failure modes are rather shorter-term than that.
I won't be surprised if current approaches can get decently good at e.g. managing a corner store or a restaurant, without scaling up to a 1000 person R&D firm.
Actually that's not generally true. Janitors at least work 12 months a year, many schoolteachers are paid ~9 despite being expected to work through them (on lesson prep &c).
It's possible to make low 6 figures, but a lot of pay is dependent upon length of time teaching and/or which district to get into. Good districts rarely have openings.
I'd say (not a teacher myself, but related to several) teaching is a lot more difficult than people give it credit for.
Just like everyone else. I have a feeling your definition of “livable” is not in line with the reality that most people face. You share more in common with a stay-at-home mom going back to school to become a phlebotomist than you think. It’s just pride gets in the way
Not enough buffer to take 3-5 years to rebuild a career is not the same as no buffer at all. Personally I have about a year of buffer which is no where near enough to meaningfully retrain.
Why are you assuming every white collar worker has got an inflated salary, or that they’ve had it for a while?
And actually no, I can’t afford to be unemployed for a year. On COVIDs cost of living it would have been easy. But the current cost of living I’m not able to save even with my “inflated” salary. And the savings I do have isn’t going to stretch as far.
I honestly don’t know how people save on lower incomes. I suspect they don’t. Which is why the debt crisis keeps worsening.
I’m lucky enough to have assets but once they’re sold they’re gone. And most people don’t even have that luxury.
Even if you've saved enough for 1 to 3 years of university, there's a risk that you won't get a job especially if you're getting towards 40+.
Drawing down savings is a very high risk move, since you could easily never recover those savings because you may struggle to get a job that has a better ability to save as much.
I don't get it. Didn't you live way below your means so you can save up money for retirement? I've got enough money saved to get away with being jobless for about a decade.
Not everyone's had the luxury of saving for years and years. Some people just got into the industry (remember 'learn to code'? not to mention recent graduates), others had to work their way up through lower-wage roles (e.g. contracting) just to get to a place from where they could hope to save. Other people have family to take care of and can't just throw everything into their savings.
Even back when I had to make do with $400 a month I was saving. Live out of a van to save on rent if you have to. There's always a way to live below your means and save.
Of course there are exceptions, but I don't expect those to apply to anyone commenting here.
A bizarre perspective - you seem hung up on the virtue of saving but totally disconnected from the material implications - even if you saved 50% of your income for a year at that time, you’ve finished with $2400 in the bank - how long is that going to float you? About a week in my case.
It’s plainly true that for a significant number of people, even those who satisfy your moral sensibilities around money, they won’t have had sufficient time at sufficient incomes to have enough saved for a significant period of unemployment to retrain.
If you burn through $2400 in a week, there are lots of opportunities to make changes in your life to make do with less. I'm currently living off $240 a week and that includes going to a fancy restaurant, so I could cut that number down even more if I had to.
You have no idea what my circumstances or expenses are, and in fact they’re irrelevant. You seem still to have a strange inability to see beyond your own circumstances.
You told me your expenses are $2400 a week. I told you that it is easy to get this number down since it's unlikely that you have to pay $2400 a week in life extending medications.
Live out of a van with children? Is that what you're suggesting people do? That's monstrous. You would rather put the weight of the world on a poor family's head than one iota of responsibility on the ruling class.
If the stock market drops 80% and stays there for longer than a year it's Great Depression 2.0. Retraining won't do jack shit. There'll be breadlines, governments paying people to dig ditches and build trails in national parks.
Lately I've been just thinking of hitting ALT+F4 on life for good, as a big FU to the megacorps (not that they would care, but it would give me the satisfaction at least)
There’s so much more you can do to get back at megacorps that will hurt them more than that. 1 less person in the world to oppose them is a win for them.
I’ve never met you but based on your last 10 HN comments I think the world would be worse off without you in it.
If you jump from a bridge, all it does, is that relatives of yours will receive the bill for cleaning up the blood and remains (on top of funeral etc).
I think of doing this except for suicide (which you shouldn't do) I fantasize about becoming a monk. That's a life role that can't be replaced with AI, provides meaning to yourself and to the community and guarantees you a certain amount of happiness – probably more than most jobs.
Monasteries are wary about people inquiring because they can't hack it in the world. And because monasteries are businesses of a sort, they are concerned with what a middle-aged inquirer can bring to the table financially.
That's a very broad statement; the reality depends on the specific order and monastery in question. To be sure monasteries are wary about all who inquire, thus the usual two-year neophyte period. Beyond that some will expect more from novices, other less, but there are many stories of broken people retreating to monastic life even today. Indeed many (within the Church) expound that as a key role of monasteries, providing a place for people to escape from the world. If you just want to live a Godly life, you can do that in society; monasteries are precisely the place you go to do so outside of it.
A pithy saying in both the Catholic and Orthodox traditions is that people shouldn't consider monasticism if they want to run away from the world, but rather only if they want to run towards Christ. A significant part of the discernment process is weeding out the former.
You retrain the same way that we told workers when we decimated the garment and manufacturing industries. You'll get the same amount of government support too.
It used to be you could just buy (for example) jeans made in America for $20. I bought such a pair back in the late 1990s. There was no need to be busy trying to "support the traditional garment industry".
Nowadays you'll be spending $40 (same price after adjusting for inflation) for something definitely not made in America, and probably of inferior quality too. Since the domestic garment industry has shrunk so much, the supply chains an economies of scale don't exist anymore to buy domestically-made garments of decent quality at those prices.
We were also told it was important to help create good jobs in Mexico so there would be less desperation resulting in mass migration from Mexico to America. Good plan, right?
Well, it was, until the factories in Mexico also closed as the jobs migrated to China, Vietnam, and so forth. So Mexico got to enjoy the same problems America had, except they were poorer to begin with.
Not really sure what I should be doing on the side since AI seems to be automating a lot of work that I do as an analytics engineer. Maybe in 10 years all I'm doing is feeding in markdowns and project specs to an LLM and reading test outputs. At what point are you just a technician?
The bright side is that analytics engineer is a relatively modern job title, previously it was largely bundled in with data engineer, and before that, data scientist. So you're either already relatively early in your professional career or have adapted between programming disciplines once already.
I also (personally! I'm in data engineering for my part) feel it's one of the safer roles from an AI perspective. Yes, you'll be writing a lot less code in the future, but that's never been the particularly difficult part of your job: SQL is dead easy to use.
The difficult part is picking up all the business context and knowledge in order to work out the best ways of representing complicated and messy data for the business. AI can help to some degrees there, but if you're great at your job, your judgement is going to be built on so much business context that both can't be written down and would be too vast to feed into the model.
Sure, AI can do some of the job, but you'll always thrash it in terms of quality (whilst getting to spend less time doing the busywork of just writing your thousandth predicate of the day!). And worst comes to worst, all the business knowledge you have in your head makes moving into analytics a far more viable lateral move.
And that one is definitely AI proof. If all companies use AI for their analytics, they're going to be moving largely in lockstep with each other. Humans might be far worse, but often far better, so I'd be willing to bet that through sheer randomness that over time companies using human analytics will outperform those relying on AI.
well, I suppose the same way that other people whose jobs were lost due to automation over the years were supposed to retrain for good jobs? I doubt they had the money or years to go back to college to get a livable job either.
In the past there was always demand for "unskilled" (manual) labor where displaced workers could land to without long (re)training, and possibly even gain skills on the job. But that job marked has dramatically diminished especially in the west; there are only so many burgers to be flipped. Somewhat famously during industrial revolution displaced agricultural workers (e.g. farmhands) ended up mass-migrating to cities doing factory work under extremely exploitative conditions. And that never truly got resolved, we just brushed the issue under the rug by pushing manual labor to third world countries where the problems could remain out of sight.
In my experience, AI has given far more leverage to software engineers than anyone else, because they have a feel for what computers can do, what's easy, what's hard, what's performant, or poorly specified, etc. Therefore I think retraining is mostly about being willing to experiment with AI-assisted workflows and not being too tied to old ways of doing things. As long as you are focused on the problems you are solving, I think you'll do okay.
What I'm more concerned about are the younger generation where all the learning tasks have been automated. I have faith that the strong, naturally curious engineering types will still learn to do great engineering, and honestly without some of the mental blocks and calcified assumptions that us greybeards have. But how do they get in the foot in the door and the reps needed to develop senior-level judgment is another question.
In some sense we need the current mania to die down and the long-term maintenance implications of heavy AI assistance to become more apparent. Basically I think we need the AI bubble to pop and then 1-2 years for things to stabilize, and then we'll recalibrate on hiring expectations and best practices for AI usage. The current AI maximalist and doomer views I think are both basically driven by capitalist incentives and wishful thinking from investors and a reality check is going to come in some form (whether technically, socially or politically).
Because I've been looking for a new job for several months and all of them either require significant retraining or don't pay nearly enough to maintain even a similar quality of life. I don't just mean they don't pay enough for luxury. I mean they don't pay enough to live in my community without a roommate.
Many of us on this site have worked directly on projects where the efficiency gains ultimately reduced or eliminated positions. If you want to exit the development field, I think the answer is probably the same as for the people whose jobs we helped eliminate.
The internet has made it easier than ever to learn new skills. You don't need college unless you're targeting a career that requires, or expects, a degree.
Maybe there will be an inversion. Lots of white collar office jobs can be partially automated, or have their efficiency greatly increased by AI. But we still don't seem to be near widely deployed robots that finish drywall
There may be an inversion as white collar jobs sink, but not because drywalling is going to get better. All of my family work in trades and it's a lot harder and less stable than we office workers like fantasize, as much as I love the ending of Office Space too. It's cyclical and sometimes there's not enough work. Cost of materials is volatile and can eat into your income. And it's skilled. Programmers tend to think everything else is easy and if you can teach yourself React then electrical should be a no-brainer. That's to say nothing of start-up costs. It was incredible that we coders could bootstrap six-figure careers with a $4000 MacBook Pro. Trades workers need a truck, you need power tools, if you're not loosey-goosey then you need insurance. All of that costs a lot.
That's for anyone joining the trade. Out-of-work middle-aged programmers with soft keyboard hands and little or no experience hanging dry wall? Why pay them any more than the 24 year old guy. AI may automate us out of our desk jobs, but the downward pressure of us transitioning to trade jobs, even at the bottom level, will only make those jobs less lucrative.
You think the drywall company is going to hire you? Why would they hire someone in middle age with no experience when there are no shortage of 20 year olds who already know the trade?
How can you not see that this time its actually different? The very need for labor is being obviated. Every technological change we had before still required a need for labor often an massive need that justified an entire city population. That is the difference. For the first time in human history, economic output is being severed from requiring maintenance of a labor pool. Considering we are that labor pool, things do not bode so well for us. As you state, look at the relative lack of abundance of candlestick makers and horeshoe makers and basket weavers and other colonial jobs. We are essentially making ourselves extinct and we don't even realize this.
There has never been a technology with the promise of obviating both knowledge and physical based labor before. Now there is. Sorry, but this is unprecedented. The hubris to imagine all will just be swell is astounding to me.
There's an off-color joke that could be made here but the real question is the speed of change - we gloss over it but there have been major changes that took centuries, and others that were quite fast, this one seems to be closer to the fast end.
We are all going to learn the hard way, this is what the economy is like for many people. Always on the precipice of losing everything and thus willing to do anything like the capital owners want it.
Whatever cushy career path you think is safe now will probably not be so safe by the time you go to university, train, land a job in it etc. The job of the future is nurse, diaper-changer, cleaner, prostitute for elderly retirees. Welcome to the aging fascist AI technoutopia.
As someone that works in robotics/automation, it pains me to see people in creative/office verticals getting all the shit with AI these days while hard labor positions, the one that everyone could be doing a bit less because they take a toll on you (working the fields, farms, hazardous tasks) remain pretty much the same.
don't we already have lots of machines that do things like this though? Even if its human operated: excavators, combines, tractors, and other heavy machinery have reduced the physical burden involved in sustaining large societies tremendously.
Yes there are fewer hands on the job site but there are still hands at all. This is increasingly becoming not the case in the future. The goal is to fully eliminate all hands and all minds. There will be no more sponges of newly available labor. Then what happens exactly no one really knows or even wants to meaningfully try to answer, because the implications are too disturbing.
I would agree with you, if we taxed the ever-living fuck out of AI companies to enable UBI, so that the people replaced with robots did not end up homeless.
Ideologically I don't consider myself to be a communist or even leftist, so I'm confused by my own reaction as well, but maybe mandatory joint-stock where the government takes a large chunk of equity of any AI enterprise off the bat and manages it like a sovereign wealth fund is the right answer.
If you use a computer to do something, training an AI to use the computer to do the same thing just takes the computer, which was there anyway. (Yeah, plus training the model, that's not my point.) But if you do something physical, replacing you takes something physical to be there.
We may eventually manufacture the robots to do all these things. We aren't there yet.
I looked into this (or rather, had Google Deep Research do a 30-page report on it!) and it's still a very long way off. So long, that the AI, or any reference it found, would not even give me a rough timeline, but I suspect it is multiple decades away. The main blocker is a very severe crunch in the supply chain of critical materials, mostly rare earth elements which are essential to make those high power density + high precision motors feasibly.
Today we have a tiny fraction (like 2%?) of the global supply needed to make enough robots to replace all humans.
This is before we get to the other logistics like power and the rest of the materials, but we have a path to scale those up (except maybe Copper, that's another tough one.) Turns out rare earth elements are actually very abundant, but extremely hard to mine, very toxic to extract (which is partially why China has a strangehold on those) and we don't have a good path to scale those up.
All of this could be moot if we figure out alternative technologies for equally capable motors that could be manufactured at scale, but AFAIK whatever promising is on the horizon would still take a very long time to ramp up.
First, you make a robot that mines rare earth metals. Then construction robots. Then an automated factory for assembling robots. The whole premise is that the robots are faster at these tasks than humans would be.
If the robotics are feasible their rise should be exponential, not linear, since each robot built makes building the next faster and helps to remove whatever bottleneck you had.
Mining is not limited by human labor, we already use huge machines (and some robots) controlled by a few humans. The limitations are investment in huge machines and processing plants, and regulations on where/how you can mine/process.
You can't just have a robot start mining/processing rare earths, anywhere in the country. You need permits, and a lot of money. I guess that didn't stop AI companies ... but there isn't an "infinite" novelty/hype premium for rare-earths, like there is for LLMs.
> If the robotics are feasible their rise should be exponential, not linear
That totally does not follow. Robots can be feasible and their payback curve can still be linear or sublinear.
Take mining rare earth metals, for instance. The problem with mining rare earths isn't that it takes so much human labor. The problem is doing it without poisoning the surrounding 30 square miles. Replacing the driver of the earth mover with a robot doesn't fix the chemistry at all, nor (I suspect) does it significantly affect the economics.
Robots have been used in industry for decades already. Most jobs won't be worth investing in the robot doing it, since human labour is cheaper than the capital investment.
Worse, it was just almost all horses. There are a few in Central Park pulling tourists and that’s enough for the Pollyanna’s to point to and go “See! We’ll be fine!”
Horses where the plaything of the rich. Farmers were using oxen not horses for the most part. However the modern plow (invented in the mid 1830s - John Deere being the best known inventor, but others were inventing them about the same time) needed the faster speed of a horse to work well, but if the previous farming technology worked for you then oxen (or donkeys ... - there are several other work animals used in different areas for farming) were better.
There are likely more horses today than there were in 1900 - but now they are play things for the upper middle class living in the exurbs (commuting via car to their city job). Sure there are a few Amish (and other similar groups) making a living with a horse, but most of them are toys for the wealthy.
Do you have a citation for the horse number? My research [0] shows we have a 1/5 out peak number in ~1900 and our population has 5x since then. This means per capita horses in the country is down 96%, while absolute count is down 80%. Both are huge numbers.
Good call! They have info on general population which shows about double the number of horses I originally thought (at least in 2016), and declining at a good clip of 10%+/decade)
. So drop is still large but just a bit smaller.
The general point is before cars horses were used a lot, entire industries were based around them and our now gone. The toy factor does not compensate because of cost and inconvenience of owning a horse.
> the % of jobs not doable by AI and robotics approaches 0 over time.
Horse are an analogy. When they were more economically useful, there were about 5x the number of them. Now they are basically pets and we get comments like this suggesting the work they do is virtually meaningless.
> Are you a preschool teacher? Are you a doctor, a therapist, an artist?
We've already automated the artists. Doctors are currently a strong area of focus. In a world where AI replaces human thought, education becomes only for hobbyists. That leaves therapist.
I suppose we're going to need a lot of therapists.
LLMs have much better cognitive empathy than most humans have limbic+cognitive empathy - the therapists will also be automated - friction like licensing bodies will slow it down but not stop it.
It's not just the Scrooges in power. Previous attempts to actually make a better world often made a worse one, and not just because of capitalist opposition.
If you mean attempts like centralized communism - absolutely - and the failure mode should have been obvious from human nature: centralizing great power incentivizes corruption.
We don't fully know the failure modes of every alternative yet, but that doesn't mean powerful central governments + capitalism is the best alternative.
Interestingly, the Amish shape their lives around this sort of thing. With any technology, they ask themselves, "Does this make us dependent?" If the answer is yes, they refuse it.
They are more and more adopting modern technology. And where they haven't directly they have found ways to use it and be just as dependent while appearing to not. Go visit one and you discover they want to borrow your phone to buy things from Amazon...
This is the Chinese century, and even if China charts a somewhat different course than Western capitalism, it still believes that human beings should spend most of their days in some employment, even if it is just makework. A population with mostly leisure time is viewed as very threatening to social stability. I can see that model being exported to the rest of the world far more than I can imagine your utopia.
The Chinese social safety net, for individuals not covered by unemployment insurance, is strongly centered around public sector jobs (公益性岗位) that, if you see them in practice, are often plainly makework: the essence of the work has been mechanized/automated, but human beings will still be directed to be somewhere all day and look busy. There is no paying those people a dole while they stay home like in sone other countries.
Many of these jobs are things that improve quality of life for everyone else, too, like street beautification programmes. They do this alongside coming up with robotic means to also sweep streets, but yes, they find ways to hire people to just "make this place look nicer".
Is it really that crazy though? Or is it an acknowledgement that the people in power would think like Peter Thiel and if given the choice between ending capitalism or ending humanity, would pick ending humanity?
Not necessarily! In "Saturn's Children" by Stross, humanity managed to wipe itself out, but post-humans (evolved from former, ahem, entertainment androids) are doing just fine across the Solar System. There's a plot point where one of them discovers that some corporation back on Earth inherited full property rights to her body due to an old pre-human-extinction EULA, and she'll be recycled for parts if she ever sets foot in their jurisdiction (I'm paraphrasing from memory).
All the rich people become bankrupt or otherwise loose their political capital, new politicians are elected in that owe more to workers and thus enact anti-capitalist policy to win over the loyalty of workers.
That or, as rich people loose more and more of their political capital, workers get empowered and use direct action (such as strikes) to demand the end of more and more capitalistic policies, until there are none left. The rich get along with it because the alternative is loose everything.
Edit: And today's industrial society is quite far from being capitalist. Feudalism is more close to reality, seeing as the one most important factor for almost everything in life is whether you own real estate or not.
That's a very loose definition of feudalism. However if you take in account nepotism and bribery, which are more and more widespread around the world, maybe we get close to that exchange of land for loyalty that was the basis of feudalism.
> HN: A site full of bold futurists where it's easier to imagine the end of the humanity than the end of capitalism.
That's an underrated comment! Totally agree. So so much of discussion on this topic sort of assumes business as usual (both literally and figuratively) and that AI quickly diffuses through the economy, sucking up all economically useful activity, with humans falling by the wayside as inevitable collateral damage.
There's a failure of imagination taking place here that there's no other outcome. Capitalism is an evolved solution to how human community works, but is not inevitable.
Massive changes in technology pretty much always involve wars and a reshuffling of power. This doesn't have much to do with capitalism at all. It's more of a question of how are those that currently have power going to adjust. In the case of technology we quite often see this end up with a concentration of power.
How to maintain stability in a world filled with greedy people in power that don't want to give up a single penny is a good question to ask at this point.
If you're not economically useful, what pressure would you be able to bring against the government and it's drone-based police force if it does things you don't like? General strikes don't work in a world with no human labor needed.
Iran recently shot thousands to tens of thousands of its own people when they protested. If a sociopath were in power, what would stop them?
>There’s no quick answers, but intuitively I think we all get that the % of jobs not doable by AI and robotics approaches 0 over time.
Can we really say that? I think that might be a prediction based on the current situation and on the assumption that jobs will neither be transformed nor created anew. We don’t know what the future holds for us in terms of jobs – or whether the problem of jobs will be a solved issue. Just as the blacksmith never gave a thought to what a mechanic actually was until he saw a car for the first time.
It is clear that at the limit, machines will be more capable in nearly every way. I suppose there might be some room for a few exceptions like working around big electromagnets that machines can't get close to... Does that sound great? Is there something else you had in mind which would be reserved for humans and fulfilling?
> I think we all get that the % of jobs not doable by AI and robotics approaches 0 over time.
I also think that will be the case. However, not all the jobs are worth being automated. Say, if you are a modern-day horse-riding instructor, I very much doubt the robots will come knocking the door to take your job. Not because they can't, but because the headache is not economically worth it.
It'd be uneconomical today but I don't see any reason to believe it will always be that way. Any time you want a high standard of living for people it creates an incentive to eventually make a cheaper way of doing what those people make their high standard of living with.
About the only things I can think to guarantee will have some constancy are "for the sake of being human" type things. Much like buying handmade bowls is something you can do out of pure preference/taste/ideology even though it's cheaper to buy massed produced ones.
I can see that going away too. Eventually the average horse rider is going to be someone of stature of a saudi prince at the rate horse ownership coupled with inequality is going. They'd probably be all about buying the Gemini horse teacher over learning from some lesser birth stableboy. I bet eventually all tutoring from a human is seen as some lesser ick by the rich. People already think as much about you today when you talk about say not relying on claude code. It is all about optics and virtue signalling pedigrees at the end of the day.
In the long run, we’re all dead anyway. (Unless something drastic changes.) That means how long it will take for robots to be practical matters. We don’t know how long it will be, but driverless cars have been coming for 15 years now and they’re still unavailable in most places.
People are more adaptable than horses and tech solutions usually require people to do some of the work. Dishwashers need loading and unloading. My robot vacuum still needs maintenance. Self-checkout means you do the scanning.
Amazon wants to get into drone delivery and there are cases where that might work, but the way news articles describe it, it’s not going to get a package onto my porch.
You seem to be living under a rock. Yes driverless cars started in 2009, but in 2026 it is very much here. Waymo recently crossed a milestone of more 500k trips I think, we will have millions of driverless cars in the coming years (we’re at 10k right now)
Waymo works in a single digit number of cities with extremely rare, highly favorable climates, and still operate only on fixed pre-mapped routes. And it's taken 17 years to get that far.
We are still a long way from a self driving car being the default type of car
There's always exceptions. To not beat a dead horse(hah!) in Texas you can still find people on horseback in the drive through line at Starbucks from time to time.
For a while waymo didn't even serve the hillier parts of where they were deployed. These roads are challenging conditions for even a human rideshare driver, when you are down to about a cars width of winding road, with a moving truck trying to pass in front of you, and a dump truck trying to move behind you, cars increasingly piling on either end, no way to turn around or back out, everyone increasingly honking and cussing as if that would lubricate things. It is kind of hilarious in a way seeing how quickly things effectively blow up in those street conditions with just the perfect storm of two slightly larger vehicles passing at once.
Now you can imagine a waymo in this mix how it would just be on its knees in a situation like that. How is a waymo to interpret "fucking move already you dumb fuck" into any sort of meaningful instruction?
I'm reminded of that story about activists in SF requesting a dozen waymos at once at some one way street and creating a localized denial of service attack essentially.
No. The horses were made redundant and no longer exist in such quantities
So, uh, maybe don't dwell on that too much.
Actually, dwell on that a lot. It's probably the only instructive example.
The rich and powerful only suffer us because they need us. They cannot assemble every lead painted doodad by their own hands. If they could, they never would have hired anyone. So what happens when they no longer need as many people?
Society is already far too constructed around catching some of the wealth that just spills off of the people who own it, by being in their orbit. That activity is not actually productive, but since the only people with money are the ones that have a lot of it (K shaped economy), there's just not enough wealth left in the bottom part of the market to support anything else.
So long as they can keep us fighting amongst ourselves in meaningless culture wars. If there is ever a broad alignment of populists spanning left to right, it could get spicy.
There are more horses today than ever... Oxen got the short end of the deal - almost nobody uses oxen even though historically they were the backbone labor of the farm, while the horse was for the right.
The only data I can find ignores the majority of horses today - recreational riders. If you can find a better citation I'd stand corrected, but everything I can find counts race horses and horses used for farming.
reality tv star on something like real housewives. I think AI/robotics would have a hard time being that irrational and illogical consistently. That behavior is pretty unique to humans.
I feel like Neuro-sama is essentially already that, but for the Twitch / YT crowd instead of television. It's trivial to train a model to maximize engagement (as one can see gestures broadly), so I'm sure if we can get robots looking nice and human, we'll have Futurama's All My Circuits in no time.
Though I guess there was a token human on that show, so maybe you're right!
The last job is going to be hairdresser, because billionaires have wives (not to assume all billionaires are men), and those wives will want a human to gossip to.
Amazing. I bet this can really help thousands of programmers who are simply not needed, because companies decided to replace them with AI agents. These philosophical ponders are very comfortable to do when your ass is safe. Most of people don’t have this privilege.
> what if what we loved about this job was never only the code?
It never was the code. The code is how you solve problems, make stuff happen. The code is a tool. It’s people’s brain that is of value. People learn how to use these tools the best they can. And now they are told that they should become mere supervisors for the AI agents. Simply by trusting the output and QA-ing the result. Because no human can keep track of the AI’s generated mess.
My son wants to enter the software profession. He isn't just immitating his father, he really considered different careers and really likes this one. He is capable -- had nearly a year of experience working at a tiny company building a high performance database in C before they had to shrink; not quite enough experience to qualify for "experienced" jobs. He is even comfortable with using AI to build code.
But he simply cannot find work. No one is hiring entry level software developers (especially not those who are better at designing code than at networking colleagues). It appears that he will be driven out of the field because of the disruption AI is bringing, and I morn for that.
I don’t particularly like the horse to engine angle because it doesn’t capture the issue well.
When we automated physical activity people that were crowded out moved into knowledge work. Now we’re automating knowledge work so people move into ???
Leisure? Become a useless class? UBI? Something else entirely?
I agree with this too and I find this article extremely depressing.
The AI "revolution" as described here is the wholesale outsourcing of vast swaths of thinking and decision-making to a single entity. A project coded with Claude is a project coded by Anthropic. A business whose decisions are made with ChatGPT is a business run by OpenAI.
Is that what these people really want? To become a conduit through which two or three companies get to exert control over everything in the world?
The knowledge economy was already on the way out in favour of the performance economy even before LLMs. LLMs have also shown that they can produce in the performance space, of course, so it remains to be seen how valuable humans are in that position but so far it would appear that many are willing to pay for a "human connection".
Plenty if not all people put more value into handmade things, but they can't afford to have everything handmade. An IKEA coffee table is $15. Handmade one from a woodworker is $400 - $1000.
Absolutely. Hence why manufacturing died long ago, and why the knowledge economy was already on the verge of death even before LLMs, and why we were already moving to the performance economy: because in the performance economy the consumer and the buyer are able to be disparate entities, where the buyers can have deep pockets even if the consumers don't.
In the performance economy it doesn't matter how rich the consumer is, but what the consumer wants to consume does matter. Because the buyer and consumer are disjointed in the performance economy, the consumer isn't swayed by cost. However, if the consumer prefers AI performances, which they quite easily could, then humans are in trouble. But so far it seems human performances remain what the consumer wants to consume.
(Technically manufacturing and knowledge work could have established the same separation, but since the bright minds went to performance nobody thought to before it was too late)
You mean my meat proxy skills while wearing a Ratatouille machine-vision head camera that tells me what wrench to pick up and how to crimp the Pex rings?
The solution is probably a trash bag with a mini forklift compatible lift rings, an optical center mark, a ratcheting closure, and make a raspberry PI do the computing millions of dollars of tooling to make a robot compatible trash bag but it might be worth it, deploy the bean counters!
Plumbing is kind of a hard job, definitely not fun for anyone above middle age. And given it will pay minimum wage, I think I'd just take a life of leisure instead.
Sure, but that is today. When AI decimates the knowledge fields, a whole lot of people are going to be looking at different career options. There will not be any shortage of people hoping to make money in the trades, and with that surge in supply will come a collapse in prices.
> There will not be any shortage of people hoping to make money in the trades, and with that surge in supply will come a collapse in prices.
And the worst case downward spiral will involve not only a surge in supply but a drastic drop in demand to go along with it.
All of those previous "knowledge workers" scurrying for the trades not only become more labor supply, but without their cushy jobs they will no longer be able to afford to hire trades people to prop up the demand side.
Apologies, I'm with you. I was speaking to the hypothetical, not trying to suggest it is inevitable. I think it is uncertain how this plays out, but I am an optimist and I think it ends up a net positive.
It is very plausible that LLMs never get to 100% replacement for knowledge workers and the winners will the people who use both, leaving behind the people who thought all they needed to do was cut costs by doing existing workloads without humans. It remains to be seen if the constraint on quantity of knowledge work was the people or the work.
It remains to be seen, but a reasonable argument against this is found in the progression of chess AI engines. After Kasparov was beaten there was still a period where the best humans working with machines were still better than the machines on their own. Eventually (around 2020 I think?) this was no longer the case, having even the best humans on Earth in the loop was a net negative. There are reasonable arguments against this, but I've found them mostly unconvincing.
> Why would it not? The economic incentives make it inevitable.
Citation needed. Again, I wanna emphasize that I'm not saying it won't happen. I'm just saying that it's not a given that it will.
There is very little sign that current models can do anything but (at best!) augment most knowledge workers (with some exceptions). And there are very reasonable open questions about whether it's economically, and ecologically, sustainable to continue development at anything near the current pace. There is, in fact, plenty of indicators that even the current SoTA is the result of a financial overexertion.
I’ve been coding for over 20 years. I may not be the perfect coder, but for me, the real goal has always been solving problems and coding is just a side effect. If I could solve an operational problem with Excel, I would! Code, once created, is a liability I have to carry.
So I’m embracing AI-assisted coding with open arms. It’s not that I don’t enjoy coding. it’s that this approach lets me solve problems faster than ever before.
Ok but you do realize that AI can also solve problems? The thing is it's not just taking away "the coding" it's also taking away the problem solving (of course I'm not talking about all problem solving).
That's the real risk here is that we are getting to the point (probably already there for a lot of stuff) where you can just point it to your already existing codebase and tell it your problem and it can basically do it.. I mean I've tested that in practice, sometimes I'll have a clear direction of where I want to go or what I would do.. but I don't really give it any hints of how to solve the problem, and many times it comes up with a similar result or at least one that also works completely fine.
So I think part of the issue is that people like you and me I feel like are going to be so much less needed because an AI can just do it instantly for basically free.
Anyway, I know this is a slight hyperbole.. but it's also not. It kind of sucks to realize that a simple prompt can replicate your work 95% of the time and do it basically instantly.
Currently we have C-level people saying "I want a website that sells widgets". Then product managers make a plan for all the pages/functionality on that site. Designers style the site. Managers divide up the work into smaller tasks. Tech leads steer a group of software engineers, and software engineers do the actual implementation.
2 years ago AI was a novelty and essentially just "fancy autocomplete". Today a skilled software engineer can use it to mostly one-shot the whole website with a detailed technical prompt and then iterate on it from there.
What happens 2 years from now? If progress continues, I expect we'll start walking up that chain of command. First you eliminate the software engineers and just have the tech leads do the prompting. Then you eliminate the tech leads and have the managers do the prompting. Then you eliminate the managers and designers and eventually you end up with star-trek-esque programming where some C-level person is just telling their computer "I want a website that sells widgets. Good! Now make the navigation bar blue. Could you make the font bigger?"
So what are your plans for when the wealthy no longer need you?
I look at this and I feel the reverse: what do the wealthy bring to the table now?
previously we had to work for them cause anything worth doing that could become a business needed a bigger team. now, a developer doesn’t need anybody above him to make stuff, and can potentially work on doing things they find valuable which imo something we always strived to but couldn’t afford to. in my mind, it’s really about ideas and execution now. and execution i bet a developer who wants to make something can do better than anybody higher up the chain for the forseeable future.
sure there’s things you still need investment and a physical presence for. but i reckon there is money and careers to be made in purely software based businesses still, and i think that value will mostly be captured by individuals and small teams moving forward.
The wealthy bring the capital required. The problem is fundamentally the commoditization of intellectual labor. It will simply be a game of whoever spends the most wins.
even in software historically, larger teams/more money have never produced better software. sure, not disagreeing that may be they’ll have all the best LLMs and the hardware to use them for a while. But I don’t foresee open source weights to the public to lag that far behind, and hardware that can run them to become more affordable.
you know who _own_ the means of production more than ever? the rich. they control your labour environment. they control politics. so you're the opposite of independent from them.
i could setup a local llm that’s nearly frontier level if i wanted to. i doubt i’ll ever need to though since tokens are becoming so cheap. although i am well off, i am not wealthy. and i only foresee this becoming more affordable in the future. may be i am wrong but imo an era when capital is less important is already here for the purposes of what i described.
my point was that for anyone who wanted to make software they couldn’t due to lack of free time or funds, they likely can give that a go now.
Then I will find other problems to solve. Do you think AI will solve every problems for humanity? it has nothing to do with coding as I mentioned earlier.
Just because there are open problems AI cannot solve, it does not mean you can make meaningful contributions. AI might become more capable than you in every way. Say AI cannot solve P vs NP. You can dedicate your life to the problem and will not make progress. Similar for more worldly problems like new treatments. Or societal problems. You are the cow.
The internet enabled the massive complex systems we have today (public cloud being the largest in scope probably, but there are sprawling libraries like everything around JavaScript, modern web browsers themselves maybe).
Without the internet these systems can not be maintained, we need all information about them available instantly. We also need people sharing how to solve problems at a global scale which the internet enables.
Personally I think LLMs will just allow even more complexity. This will create more situations that need human intervention to solve problems. And also new problems we have never seen before.
Hah I have no idea. If we reach AGI in 10 years we will still have problems. Otherwise we all would be enlightened by then, right? Until then I think mental health would still be a problem. Maybe I should get a therapy license now.
> Then you eliminate the managers and designers and eventually you end up with star-trek-esque programming where some C-level person is just telling their computer "I want a website that sells widgets. Good! Now make the navigation bar blue. Could you make the font bigger?"
Those C-level people won't want to do that. Plus in a competitive economy everyone else can do the same so there will be competition to remain relevant.
1 website won't be enough, you'll need 10 and each site needs 10 times the features or your buyers will look elsewhere.
At the end of the day the C-level person delegates to a techy person, it's not the C suites job to care about font sizes or navigation bar colours.
> Ok but you do realize that AI can also solve problems?
Why have companies with infinite amounts of AI and expertise not deployed comprehensive end to end solutions for business problems?
Every solution they've come up with is "here's a tool. The solution to make money from it is left up to the reader". Why would they leave all that money on the table?
What you said is valid but I think you are still constrained in thinking that the code is the solution, though. I wish I could word it a better way. Maybe we need a word calculator /s
It's not right a lot of the time (yes, even the latest models) based on the problem you're trying to solve for your users. (That's one thing we can keep doing).
It doesn't always know the right thing to fit the need, so you can now figure out what that looks like and design it appropriately (and try 10 options in production and see which one wins!)
But also you can just do a lot more work. If the prompt can replicate 95% of your work, now you can do keep moving forward faster and do a lot more work.
This does mean, of course, that there will be less of a need for software engineers that are only capable of reasoning about the edges of applications or one narrow part of the whole discipline will get squeezed out. If a computer can do the job, it's now less valuable of a skillset than it used to be in the market, and that hits wages hard if your only contribution to the market was "I know how to create React components when someone tells me what is in them, and how they need to interact with other components"
Happily, despite the amazing productivity of AI, there is still so much absolute CRAP software out there, and so many different places we can apply better options. Bureaucracy and poor management will continue to create new opportunities.
My grandad ran a small letterpress shop and when phototypesetting came in he didn't retrain, he just did smaller and smaller jobs until it was wedding invitations.
“Retraining” is mostly a myth. Former skilled workers just downgrade to service jobs, pushing out younger, less skilled workers in the process.
The NYT touched on this issue in a podcast about NAFTA: https://www.nytimes.com/2024/10/08/podcasts/the-daily/americ.... AI is going to be NAFTA on steroids. Politicians will be able to point to growing top-like GDP to say it was a good thing. But it’s going to brutally hit a lot of people and most of those folks will never regain a secure economic footing.
Retraining for typesetters meant becoming "prepress" guys, and going from the wide array of tools and hand skills they used to need to Photoshop, Illustrator, and digital imposers and plates. In the process, the job went from a middle-class job to a minimum-wage job, and the bosses who used to respect you (after the union that used to protect you disappeared) became fly-by-night scumbags who treated you like dirt.
As far as I can tell, the entire business for someone who was good at running a press is fancy wedding invitations and business cards with raised type. I am speaking as someone who waltzed into (and later out of) prepress because I knew the digital tools, made no money, and talked with those guys a lot about the past. This discussion would be while we were printing wedding invitations and fancy business cards.
It's simply a different, low-status job now. Retraining is no different than training the first time, because if you want to make a living you either have to somehow be better than the people who started with the new tools (and the new tools don't require as many people) and maybe get into printing packaging or manuals, or you need to do something completely different, and compete with people just starting their careers.
Going to college again is not a trivial thing. The mortgage will still be due.
I'm not sure I can quite agree with that. I'd say that spreadsheets definitely reduced the need to manually calculate things, and calculators before that, and the abacus before that.
I basically agree with your "higher level reasoning", but not your statement that we previously only automated muscles.
> Today, the number of people employed purely for their muscle is very small.
I also believe we underestimate how much that costs us too. Yes, it is good that fewer people are breaking their bodies on brutal physical jobs.
But there are many people in the world for whom their physical strength and dexterity are perhaps their most valuable asset. With the loss of those jobs came a loss of an avenue of meaning for them.
I could be wrong, but I have a hunch that the whole manosphere alt-right conservative thing going on in the US is driven in part by a generation of young men who don't find intellectual work to be appealing but don't see many alternatives out there either. It's telling how closely Gamergate is tangled up in that world and how many games are basically simulations of heavily physical prestigious roles.
A recent Onion headline really hits the nail on the head: "'This Here Is Probably Our Bestselling Love Seat,' Says Man Who Would Have Been Powerful, Revered Warrior 4,000 Years Ago"
Compilers automated higher level reasoning, at least to a small degree. LLMs themselves are basically just compilers with a much, much, much, much more expansive set of built-in language constructs.
Can you describe the last time you asked your compiler how you should best approach an interpersonal problem, to research an investment thesis, or to figure out an org chart by analyzing communication patterns within teams?
People aren't just using AI for writing code, but for all of their decision making. AI is far from limited to writing code.
> Can you describe the last time you asked your compiler how you should best approach an interpersonal problem
Sure. I wrote out specification that described the problem. It compiled that into a description of how to deal with it.
> People aren't just using AI for writing code
Compilers have never been about producing code in particular, though, even if that is where they are most commonly used. A LaTeX compiler, for example, outputs books. To think they are about code exclusively is a grave misunderstanding of what a compiler is.
> Compilers have never been about producing code in particular, though, even if that is where they are most commonly used. A LaTeX compiler, for example, outputs books. To think they are about code exclusively is a grave misunderstanding of what a compiler is.
And what, pray tell, are those books output by TeX? That's right, PDF or PostScript code.
I don't think I knew how to approach the problem. There would have been no benefit in using the compiler if I did. Kind of like, if we want to talk about code, many coders don't know how to approach making a hashmap. They don't need to. The compiler figures it out for them, as it did me for my problems. I'd call the input I gave a specification of the problem, but words are in the eye of the beholder, I suppose.
If we take "write the specification of how to approach the problem" at literal value, you might be narrowly thinking of a low-level compiler, or what sometimes gets called a transpiler? However, compilers encompass a much wider spectrum than that. Generally they enable users to operate at higher levels of abstraction.
I think you're talking about more than just a compiler, though standard libraries are often distributed with the compiler. Take Rust for example, it's not like the compiler has special integration with hash maps, it's just that the standard library has a standard implementation and the compiler lets you refer to that implementation without having to repeat it yourself. In a way you're just including parts of another spec in yours. Specifications do this all the time, for example by using English. You don't have to define all the levels of abstraction, only the ones that matter.
Even so, I'd argue that you still have to define how to use those abstractions rather than only the result you'd like to achieve with them. Sure, maybe the abstraction has a ".achieve_the_result()" method, and calling that method is how you achieve it, but often there's a bunch of stuff you fit together to achieve a solution to your problem and the compiler doesn't necessarily do that for you.
I think there actually may be some compilers that get closer to this, for example some problems can be expressed quite elegantly in Haskell or APL and the compiler will do a lot of black magic behind the scenes to get you a performant executable (somehow). But for most traditional compilers you are still going to have to define how to approach your problem, even if only at a higher level thanks to libraries.
> Sure. I wrote out specification that described the problem.
The LLM usage I see from most people has the LLM write out the specification, asking clarifying questions as needed. If you're using LLMs as you describe, you're using them ineffectively.
> The LLM usage I see from most people has the LLM write out the specification
If the compiler is writing the specification, what purpose does the person serve? That is not what I'm seeing out in the wild, though. Perhaps what is happening in research labs aside, compilers, even those implemented as an LLM, don't do anything until the user specifies their intent.
In fact, you literally recognized that in your earlier comment, asking about my action to specify my problem. Not sure why the story is changing now? There is probably a day where the systems solve your problems before you even know you have them, but we're not there yet. The technology hasn't advanced that much in the span of a couple of hours.
> asking clarifying questions as needed.
Sure. Even the earliest, most primitive compilers would seek clarification if there was something that it felt required more information. That's expected.
> If you're using LLMs as you describe, you're using them ineffectively.
LLMs are just compilers so they can't do anything else. Maybe you're thinking of some technology that is layered on top of a compiler like an IDE, REPL, or something to that effect?
> Sure. Even the earliest, most primitive compilers would seek clarification if there was something that it felt required more information. That's expected.
Can you tell me what the last question GCC asked you was, and can you rate its coherence in comparison to your colleagues questions?
This is a tired, idiotic analogy. For one, if LLMs are compilers, they're terrible compilers because they can't be reasoned about the way one wants to reason about a compiler (but I assume you have a weird take here since you claim compilers are non-deterministic).
But more importantly, if you insist on treating LLMs like compilers, then the source language – natural prose – is the worst programming language ever invented. Coding the world in brainfuck is a better idea than where you lot is suggesting we go.
My key thought is this, many of us necessarily work in low-purpose, boring corporate development jobs. The code was the only thing making it enjoyable, because problem solving was the skill we enjoy exercising and you get that in spades with programming, even if the actual business is soul sucking. This is the reality for many professions not just software. The second part of that reality is that we have to work to pay the bills, and stability is an important pillar that many families rely on. I mention that to answer the inevitable question "why would you work there then?". There isn't an abundance of fulfilling purposeful jobs unfortunately.
That problem solving aspect is being replaced by AI, and that means anyone in a position like that is losing the piece of the job that they enjoyed. The more apt analogy is that the farrier worked in a farrier factory, and the horses were actually mass produced statue horses with no purpose but making money. He still enjoyed working with his hands, but now they have replaced that part of his job with a button to press. Now he presses the button in the horse statue factory. He still has a job but now no joy, and eventually they will automate the button too. The outlook IS bleak, and many of us will move on to totally new industries and that's fine. Some will stay because they have a rare job where they have agency to problem solve on the layer above code. Many still just press the button to feed their family.
I agree with your first paragraph, a majority of corporate programming jobs probably suck. However I'm not convinced that AI will reduce demand for these positions to in a major way. It will instead change the required workflow. I've got a couple of industrial revolution points I'd like to share;
The Jevons Paradox addresses demand concerns. Software is now theoretically faster and cheaper. More projects are now financially viable, since total unit costs are down! The other side is the Automation Paradox, where now even more specialised knowledge is required to fix and maintain AI systems.
Your disenfranchisement is valid, however fortunately (or unfortunately?) do-nothing soulless corporate programming jobs have a good chance of sticking around, if in a changed way.
I think this is true, but I also think that AI has made the mental burden of putting up with a less than ideal job much lighter. I am no longer mentally exhausted at the end of a long working day, yet I have made a meaningful difference in terms of technical progress. And, at the end of the day, I am still bursting with energy about exciting personal projects I can now take on and make real progress on because I'm not dead at the end of the day.
Not sure if this fits, but the mechanic, rather than performing hard manual labor mounting horse shoes on horses, now just has to read software error codes during the day, and gets to enjoy driving his sports car in the evening.
Back in the '70s my dad used to tell a joke about his dad. He said that my grandfather had a job that supported his whole large family while my dad was growing up. But unfortunately technology caught up with him, and he was replaced by a device just this big, and he held his fingers apart to demonstrate.
Then he said the real disaster hit: "And then my mom got one."
Grandma was there when he told it more than once, and thought it was hilarious.
It usually came with another joke about the time he was caught with his pecker in the pickle slicer. He was fired, and so was the pickle slicer.
In the seventies you were expected to tell jokes like that and allowed to laugh at them.
I think the punchline is that workplace romance usually gets both parties fired, regardless of who initiated. Hence, doing "things" with the pickle slicer means the pickle slicer gets fired for participating...
The first one sounds like the dad was replaced by computers. But then reveals the dad is a sex worker and being replaced by sex toys.
The second one sounds like the person was being intimate with dangerous kitchen equipment, then reveals that a "pickle slicer" is a job for a human and not a machine.
My grandpa was a shoemaker. The real change in technology left his family in poverty in Chicago during the depression. I didn't hear anyone joke about that, directly.
Jewish humor is frequently about laughing at your own pain. There's a genre of Jewish jokes created in concentration camps.
However, the one thing I feel like is missed by some in these conversations is that what's being built is a general purpose thinking machine, not a coding machine. How are we going to transition from writing code to solving problems when the machines are solving the problems?
You can't outsource understanding, and to the extent LLMs can understand anything they can only learn through training. So until LLMs can truly learn, rather than bringing themselves up to speed with huge contexts, you'll probably still want some people who understand what it's doing.
I think Navier Stokes is a brilliant case study for all that it didn't do. An AI 'solved' a famous problem, of a kind that is impossible to cheat, yet the reaction is incredibly apathetic. Why? Because it apparently failed to do the thing people were hoping it would, but what exactly? Of course it also manages to fumble the delivery by antagonising the people most interested in the result, but still.
The honest AI boosters do claim we won’t solve problems anymore, we will live animals in a zoo, overseen by machines of loving grace.
The retard AI boosters, tell us to be AI first, learn this harness bro, loops are yesterday, now it’s graphs, use 100 agents a minimum or be left behind, actually world will require 100X more software engineers.
Why would we still be alive? Once we have advanced enough AI inside not only computers but robots, the rich will no longer need us. They will have the robots farm the land, clean their homes, fly their planes, etc. We will then have no income and therefore no way to feed ourselves. Most people do not own enough farm land to keep themselves alive.
In the past, such imbalances of power were resolved because the rich still relied upon the poor for everything (security, food, etc). I think we're approaching the point where even armed insurrection is no longer an option with the advances in automated killing machines. We're not there yet, but it's coming.
Other than Uber and Uber eats, I can’t imagine myself doing anything other than those. I might get a security guard certificate and guard empty halls during the night, which quite suits me.
Not a handyman, not suitable for physical jobs, I think I’m cooked! This is the fear that is always on my head.
What makes you think those jobs will still be available?
Uber : Getting replaced by Waymo
Uber Eats : They've been experimenting with ways to replace humans for this. I think the most likely scenario isn't those silly robots on the sidewalks but rather just a Waymo. Have the restaurant send someone to carry food from the restaurant to the Waymo idling at the curb, and have the customer come out to the street to pick up their food from the Waymo. We lose the door-to-door convenience but I think a lot of people would take that trade to eliminate the cost of tips and paying the delivery person.
Security Guard : AI monitoring security cameras and automating calls to police sounds trivial. There are also already "security robots" that move around so they could be dispatched for scenarios which do not warrant a call to the police.
Handyman / physical jobs : While I think these will last longer than the others, I think they won't last much longer. We've already invented all the necessary hardware. The limiting factor was the software. What happens when you take an LLM from the year 2030 and give it a pile of tokens to write the software to drive the robots to do plumbing/building/whatever.
> Uber Eats : They've been experimenting with ways to replace humans for this. I think the most likely scenario isn't those silly robots on the sidewalks but rather just a Waymo. [..] We lose the door-to-door convenience but I think a lot of people would take that trade to eliminate the cost of tips and paying the delivery person.
Amazon is already able to do deliveries for some packages by drone in some cities. This seems perfect for it.
This is my dilemma as well. I can't think of a role that won't be impacted so I don't know what to train for. When I was younger, my fallback was always going to be "truck driver" but that's going to be one of the first dominos to fall.
Also, it leads to significant lower purchase power, which may indirectly decrease profits, at some point ? ("Firing your customers" works up to a point)
Don't forget blatantly ignoring all the times something similar happened in the past and didn't turn out fine.
Sticking your fingers in your ears and singing "Lalalalalala" will in fact sometimes work for some people, and that is not a good argument for everyone to do it.
I like this article, but it took decades for cars to take over: we needed to build vehicles, build roads, build up gas production and distribution... as Bill Gates wrote, one of the differences with the AI disruption is how fast it's coming, since it's almost immaterial (except data centers of course). It may be too fast for humans to adapt like they did in the previous revolutions.
The current trend of comparing AI development to the development of motorised personal transport makes it very easy to remind everyone that undoing the effects of decades of blind "progress" once the negative externalities become unavoidable is much easier said than done.
It can replace all work, IMO. I'm doing intellectual work that replaces manual labor with machinery and computer vision systems. It's massively sped up by AI.
The grandfather must also have survived two world wars. The transition from horse to cars happened within a broader context and was greased by blood.
I find it curious that this is often elided when discussing technological transition. Sure the whip makers went away but they probably also ended up dead in a trench.
So far the impact of LLMs have been very much overstated.
Like most (all?) tools, they require a human in the loop to work as intended.
This fear of being replaced by an LLM is unfounded and irrational.
The alternatives that may cause such mass displacement like World Models and such are yet to be proven.
LLM's went from generating incoherent garbage, to weird text that seems coherent but isn't (gpt-2), to what we have no with agents passing some of the hardest benchmarks, necessitating benchmark development just to be able to keep up with the rate of progress. They did this in about 5 years.
The sheer rate of progress is what is meaningful. If LLM's agents were frozen now without any further improvement, your position would be must more defensible. But that's not what's happening, we are getting breakthroughs almost on a weekly basis.
It feels to me like the advancements are not accelerating, though, like we're nearer the top of the S-curve than enthusiasts think we are. A lot of the recent advances seem as much a harness improvement as an actual model improvement. We've made exceptional progress in the last couple years, but spent an exceptional amount of money to do it. A moon shot, basically.
Honestly, my gut feeling (worth every bit of what you are paying for it...) is that until we get actual intelligence instead of a good facsimile, we're not looking at an existential problem or a UBI-level problem, even, just a big change in tooling.
Hey I hope you are right. I, like so many of my fellow engineers, am not exactly thrilled at the prospect of my role changing from underneath my feet and the uncertainty that comes with it.
We are making steady progress. It’s not exponential per se like we used to see but in my experience models are steadily getting better. Also breakthroughs can be sudden.
"So far" is a statement about the present and past. "This fear" is primarily about the future, though a few jobs already have been affected. So it's not like there is zero evidence.
While I appreciate the point being made here, a tangential thought:
> A century later, the torments felt by my great-great-grandfather are forgotten. The garage feels like it has always been in the family. My father loved his job like few people I know, as if it had run in the family forever.
Arguably, the internal combustion engine is one of the most unsustainable and devastating technologies invented by humans. The jury is still out on whether our species will survive the environmental impact wrought by its effects. So much benefit, but so much dependence.
Dont you think silicon based AI will soon outstrip the energy it needs to run at AGI scale? Since we can already grow brain organoids (and let them spread out flat instead of being crumpled up in a ball) we could scale brain size enormously. And since human intelligence only uses 25 watts, imagine the superhuman intelligence of a 25kW brain (the energy used by a server rack). Every neighborhood could have its own superAGI. And not only that, but new brains can learn from old ones (a better approach than distilling models)
I like this idea, but I don't think it's feasible. First, neurons are slow. They fire at most around 1000 times per second, while transistors switch billions of times per second. Second, although we don't fully understand how the brain computes, it doesn't appear to work like a discrete state machine or to rely on symbolic manipulation. We could probably make wetware emulate CPU-style behaviour, with discrete states and symbolic operations, but we'd likely end up worse off than with ordinary silicon.
> Many of us, deep down, are driven more by the joy of making things, solving problems, and seeing people use what we built than by the act of coding itself.
I've definitely always enjoyed the end-result but only for the fact that it proved I have overcome challenges and slowly but steadily inched one step closer towards mastery. Both are now gone, since mastery is not required and challenges are absent except the soul draining exhaustion of reading markdown all day. Now I'm just producing outcomes for my boss, rather than my own fulfillment. The journey has become a farce, I will ride this out and make my money but I don't care about it much anymore.
That being said in my free time, as in Hobby I am doing pentesting manually for fun. Because productivity is irrelevant for my hobbies, so I can still keep inching towards mastery and overcome challenges, at least on days where my job leaves any trace of surplus energy at the end of the day.
I used to find joy in knowing how things worked inside-out. Whether it was Postgres, or browser engines, or some detail about compilers... I used to find joy in developing the little techniques that made programming NOT building on quicksand. There was satisfation in what you had done being robust.
Pardon my lack of creativity but I'm not even sure what mastery with the new tools (LLMs) would look like.
When you are so abstracted from the details and don't have to know anything, what _is_ mastery? (I am not asking rhetorically and if people have overcome this thought I would be interested in hearing from them.)
Knowing how to ask a machine questions in the right way? Playing as a manager in control of non-human developers?
It all sounds really silly and something that is difficult to aspire to (and I can't imagine explaining my job to others with a straight face).
I think I find it even more egregious because it's all built on paying rent to the LLM operators.
It's not that I don't enjoy results in their own right sometimes but I think I am definitely in the 'this is giving me an identity crisis' camp.
How do LLMs work? Yes there's a model and there's inference, but dig deeper. How can you, today, with < $50, train a small GPT-2 class model,
how do you fine tune it, what training data do you give it, how do you give it training data, what algorithms do you use for reinforcement learning, what even is RLHF? You don't have to know any of those details to use ChatGPT, just like you don't need to know how a D flip-flop works in a CPU to write some code that runs on one, but if you're looking for feeling like you know what's going on, at their heart, LLMs are still next word prediction engines, so knowing, deeply, how that engine is created and works, would be my idea of mastery of it. Karpathy has a good set of tutorials on this.
>How do LLMs work? Yes there's a model and there's inference, but dig deeper. How can you, today, with < $50, train a small GPT-2 class model, how do you fine tune it, what training data do you give it, how do you give it training data, what algorithms do you use for reinforcement learning, what even is RLHF? You don't have to know any of those details to use ChatGPT, just like you don't need to know how a D flip-flop works in a CPU to write some code that runs on one, but if you're looking for feeling like you know what's going on, at their heart, LLMs are still next word prediction engines, so knowing, deeply, how that engine is created and works, would be my idea of mastery of it. Karpathy has a good set of tutorials on this.
Knowing how an LLM is built, is trained, inferences, isn't
"mastery" in the sense that crafts people use the word.
Mastery to a craftsmen is as much a product of mechanical performance as it is knowledge. For a software craftsmen, that means knowing how to code, debug, and architect a software application by hand.
This, plus since you can't realistically run these LLMs at home, the motivation to get invested in them is not really there (for me at least)
I mean, yes I do find the mathematical background of LLMs fascinating, but it's not going to be something that I can put to practical use (much), if that makes sense?
When I would code by hand I often found I was concerned with the house of cards in my head, am I remembering the right variable names, did I forget something that would lead to an exploit, etc. Now I can direct the LLM to do the work and the skill is split between having an understanding of the interactions of the technologies, choosing tech that will be less of a pain to update/migrate. You can even put some LLM-bow grease into finding more efficient approaches catching corner cases etc. And just by asking I can have test frameworks setup, I can drill into some pet peeve and it takes less time. Now we still have lots of ops moments, like the radicel distributed git forge forgetting to use encryption for private repos, and I run into this class of problem in my own casual LLM use, but that's where our taste and attention to detail powers need to be spent in this modern developer world.
I think one thing that troubles me is that I so far have gotten quite far by 'caring about the little things'.
Now we live in a world where this feels, for the most part, unsustainable to keep up with.
When directing LLMs to code something up, it's still occasional useful to have the knowledge and experience of those little things earned over years, but at the same time 1) it seems like I run into fewer opportunities to add to that experience and 2) it seems like you can often just prompt LLMs to 'grill' a design a few times, it will probably surface some of the most important tradeoffs
Yes, the grilling part is where the LLM energy drain is for me, I haven't released any of the software I've been working on yet because they lack polish in my opinion. They may have nobel goals, but no point unleashing more slop on the world. Often when I ask for a correction something new breaks, or it reveals the next corner the model and I didn't think about. I've seen the models come a long way in a short time.
I'm tinkering and paying by the token for deepseek flash picked for its economy and using it to teach myself and form opinions on languages and frameworks that are new to me. It's hard to describe the worth of these opinions, but I think they're important, kinda like why I read a lot of HN looking for morsels of sentiment around the tech industry as represented here.
Well, I'm somewhat minded to agree with you, but they are fast enough, cheap enough and don't get bored of tedious work, which makes them passable at many things. Maybe.
I feel caught between two worlds, I know people both very 'for' and very 'against'.
The business is very keen for people to use them to go faster.
For non-trivial changes, output so far does not meet my quality standard and I tend to spend a lot of time cleaning up after them, probably more than I would have spent doing the work myself. This being much more boring, I hasten to add.
Their comments suck so I feel like I have to go in and rewrite them all.
Maybe I should just let go of the quality standard and accept that in exchange for steaming ahead at full speed, but I'm reluctant (and not just for me, there are also coworkers who I feel like I would be letting down).
I've always been either entirely self-directed, choosing my own goals, or almost entirely self-directed, being given high-level goals, and getting to choose my own means to accomplish them. So my opinions are probably different from somebody who was a smaller part in a chain.
You say that you liked learning how things worked inside out, whether it was Postgres or whatever. You can still do that. You can learn it at a high level, and then it becomes like a magic word in your arsenal. You don't have to know or remember the details of implementation, but only the nature of its costs, benefits, and synergies with other systems. You just have to remember that high-level abstraction, maybe just a single vocabulary word, of a low-level reality - the exact syntax is really no longer a concern.
The better the agents get, the less there is playing manager. Really your job now is to have and execute vision. Execution is largely a function of having a map of the high level vision and low-level realities in your head, or just having in-the-moment intuition about these things... then just speaking it.
Also, there is developing personal libraries. You may find that in your business you repeatedly tackle similar problems. You develop and refine a library that your future agents can utilize. You develop prompts that make your work more automatic and more robust. And when I say you, I mean, you direct your agents to do it.
It's not so different from what it always was, except that years get compressed into months. What was slow and tedious is now... less so.
I don't know how much advantage I have in this new world, having programmed (sometimes more, sometimes less) for 30 years. My brother has never written a line of code in his life. He's been heavy into AI for a while, and he has developed some really incredible stuff. I think he may have already outpaced me. His $1500/mo+ in subscriptions definitely help.
Did a tailor gain mastery of the sweatshop machine? Did mastery of the automated sewing machines give the same joy to the tailor or was it mind numbing meaningless activity that he would be happier to be free of.
that's a good analogy and puts into words the way i've always felt about software development. Every so often the entire industry rolls over into something new. I've been here 30 years now and, in my experience, the skill that counts is being able learn new skills. To go back to the analogy, to build something new on top of what you've built in the past before everything is swallowed by the quicksand.
I think well north of 90% of people in software engineering have job expectations that, when assisted/automated with AI, boil down to exactly that.
It's a pain in the ass to work at such a job and find challenges or things to master--your superiors don't want you to do that, there's not enough time in the day, and you're tired from LGTM-now-make-this work.
That's most software jobs, because that's what most software companies want.
Because I literally can just burn tokens instead of using any of my brain cells and the outcome is way better than what I would have achieved thinking deeply and intricately about the implementation because Opus 5.5 at xhigh implements way better than I ever could and I'm paid well and have studied for years and it's not a low-level environment at all.
Since opus 4.8 the tools have started surpassing me and now they are far gone, so implementing by hand would mean sacrificing quality and my main goal as an engineer has always been quality so even though I hate it, I must use the best tool for the job.
The information revolution has been going on for >50 years. AI is one event in a broader arc. Wait 10 years when the humanoid robots become reasonably capable and affordable. We’re in the middle of it, but still so so far to go.
> Wait 10 years when the humanoid robots become reasonably capable and affordable
We've had humanoid robots that could walk for what, 30 years? While technology marches ever forward, and sometimes makes pretty good leaps, 10 years for both capable and affordable seems pretty optimistic.
I’d agree that the word “reasonably” is carrying a lot here. But call it 20 years, no question. Look how far computers came from 1970 to 1990, that’s probably about the pace to expect, or faster.
I don’t know, somehow the pace seems different, these transitions felt more like the last generation retiring, and the next generation doing something different, instead of it happening to everyone at the same time.
A less misanthropic alternative: The technology required to replace the person is much smarter and more adaptable than that which was needed to replace the horse. Because of this, you can end up in the horse's role in the story without having to be as dumb as a literal one so long as it's this other lever which moves instead.
The job of a software engineer has always been to turn a real-world idea or problem into a precise specification a computer can execute.
That specification used to be assembler, then higher-level languages like C and Python. Now it might be detailed Markdown files that AI turns into code.
So our understanding moves from the C/Python level to the Markdown specification level. We still need to understand exactly what we're building and why.
The hard part was never translating the spec into machine code. The hard part is defining the right spec. That is still the job of the software engineer.
It is more like, "I'm a carpenter, and electric powertools that can also operate themselves, and run the sales department, and control a robot to change a lightbulb, and manage the entire company, and surround it with armed guards it also controls, just came out. I'm screwed."
I just wanted to point something out. I hear people say the AI revolution is the biggest one since the industrial revolution. I think it's the biggest revolution since the one that just preceded us, the information (computer) age. So we're still going fairly in order revolution wise.
The biggest revolutions since the Industrial Revolution were probably trains first and everything related to electricity later. Telegraph, radio, lightning, TVs spanning over about one century. And advances in health care, agriculture. But it's hard to tell without having lived through all those changes.
Then computers of course, but not as impactful as the other ones. And cars and mobile phones (mobile dumbed down computers) have been more important than AI so far IMHO.
I am already seeing more companies hire people who are developer-adjacent, especially now with AI-assistance.
I don't think software developers are going away, but likely headed more in-house, as each company can cheaply have bespoke software made and maintained.
I think the big problem is that nowadays, you can't simply switch professions any more, A.I. can do just about anything white collar, and there's only limited amount of blue collar jobs to go around.
Having managed programmers for more than 30 years, there are a LOT of programmers who just want to code. There are also a lot of programmers who went into programming because "it was a good, well paying job".
Doesn't really take much away from the article but a significant percentage of the developers who are unemployed are not losing their passion, they are worried about what job they will get to pay the bills.
I think the analogy of moving from horses to engines breaks down when going from code to AI as AI can quite plausibly work on itself. It's a fundamentally different game we're playing this time. And there were as many engine jobs as horse jobs (maybe more). The whole point of AI is super human productivity. Even the idea of just using AI means you need less humans because one human with AI can out-produce multiple humans without. It's very difficult to imagine a future where AI can do every job and things don't turn into a sci-fi dystopia.
My grandad on one side of the family was a market gardener. He didn't become a farmer.
My great grandad on the other side was a lighterman[0]. He didn't become a ship captain.
Sometimes professions really are just gone. In both cases in my family it was due to consolidation of resources and concentration of capital/power. Technology was just a catalyst for that.
> Be willing to let go of the how to hold on to the why.
Without the how, I have no interest? The 'why' isn't the end result. I don't read a book so that I can say I've read the book, the purpose is the reading, not the state of 'having been read'.
> But it makes me wonder: what if what we loved about this job was never only the code?
This exactly. Look, I "love" code but only because of what it allows me to do and there is plenty of code/coding that I don't love. Not to be too hippy-dippy but I'm a builder, I like to build things. Build things for myself, build things for others. The code was _always_ a means to an end.
Sure, it was cool I had this skill that many people didn't have but I also used to have skill in managing the windows machines and things like the registry. When I first switched to macOS I felt lost, I had this "skill" that was no longer needed but after a while I realized that was a good thing, I wasn't wasting my time fiddling with crap that was just getting in the way of _using_ my computer.
Similarly I may mourn the loss of writing code but at the end of the day code was just a way for me to build, if there are better tools to build (LLMs) then I'm all for it.
I too love building things, but by building things I mean the act of actually building it. Thinking of a problem, devising a solution and then executing the steps to solve the problem, if that doesn’t work, debugging and trying another method and so on. I am not and will never be a fan of buying things, where I just ask for something and it gets made which is what you’re a fan of if you’re honest to yourself.
> Many of us, deep down, are driven more by the joy of making things, solving problems, and seeing people use what we built than by the act of coding itself. After all, many of us decided we wanted to be developers before learning to code, just to create video games or websites. The desire came before the code.
This is a big disconnect that I think has been at the root of many different reactions to LLMs lately. Here on HN I see a lot of "I hate what LLMs did to programming" and equally a lot of "LLMs are awesome, I've never had a better time". I think the disconnect is between people who build software for the end result (having users, the joy of completing a thing, etc) versus building it for the process itself (architecting, coding, debugging).
I'm mainly the kind who enjoys the process. I think of an awesome startup job I had that solved real problems for real customers including some really big firms, but I can't honestly say I ever cared about those problems or the customers. I loved the job because it was technically challenging with a lot of freedom, and the customers were great because they were a source of new interesting details to solve, as well as confirmations that the previous solutions worked.
As such LLMs make me sad because they effectively replace a lot of what I like doing. Doing a refactor to support a new feature, digging through the code to try and understand a bug. One of my strengths as a developer, I think, is that I am fast at familiarizing myself with unfamiliar codebases. That's also a somewhat redundant skill as Claude can answer "how does feature X work in this code" in a small fraction of the time it'd take me.
A few years ago (before the LLM craze) I moved to game dev as my main job, which makes a lot of this easier because I actually care about the end product and the customers (players), so that's a whole new layer of satisfaction in addition to the usual technical one.
Which isn't to say that it's all bad. I've used Claude to get a few things done where I mainly care about the final result and not the process. There was one such project I wanted to get done, I'd had the idea and the data for five or six years, but I found that the implementation would be time consuming, often manual work and much of it involving the kind of tech I don't really enjoy. So I never worked on that but last month, I did the whole thing in a few days with Claude and got exactly the result I wanted.
I've been seeing things the exact same way. There are those that want the end product as fast as possible and there are those that love the process and the struggle of getting there. The former group is the ones telling the latter group that they'll be left behind.
On the one hand, it makes sense to think of this as just another revolution as it keeps things familiar, keeps the ground underneath us intact and lets hope it goes this way.
On the other hand, if long horizon independent problem solving with almost zero human intervention or support is realized in these models 1-5 years down, then it will lead not to new tech, but essentially a new species of workers that can replace humans at all tasks, In that case, humans wouldn't really have the upper hand as an economical resource to get any work done.
Why is noone talking about the elephant in the room : at current LLM efficiency per W it's still around 1000x the one of our brain; interpolate and to replace us all we need to harness more than what the sun delivers...
Not a single comment about efficiency or consumption, only marketing stuff and bad analogies, not what it used to be HN... Where are all the nerds ?
That's the problem: if everything stayed at the current level of development, we would be fine. There is no reason to think that will be the case. Intelligence per watt has been rapidly improving [0]
> 1000x
Coincidentally, that is exactly the efficiency gain promised by ASICs like Taalas [1]
> Because, yes, we love to say AI is the biggest revolution since the Industrial Revolution, and that we’re going through a transformation unlike anything we’ve seen before. But that overlooks the farmers who watched the tractor roll in, or the farrier who watched the car arrive.
How does that overlook the farmers? They're alluded to right there in the prior sentence. I don't get this.
The mechanic analogy is rather useful. Today a mechanic is in a more squeezed space than we are, having to learn EVs and electronics, or accept to go out of business within the next 10 years.
Having worked with a few combustion engines I have no clue about how to repair an EV.
A fine example of how new technology is not better, since it closes the door on too many people.
sorry I think this is a false equivalency. This isn't the career path of programming getting replaced by a new technology. This is technology replacing career paths.
We're all familiar with the term "knowledge worker." What happens when that field is broadly removed? It will be a reverse-trend of telling all ex-coal miners to become software engineers. Except here's the rub: going from a software engineer to something requiring a human only for their dexterity is typically a massive earnings decrease. So, in effect, as Brian Merchant and others put it, this is not just a shift but a massive plan of wage theft.
Looking at it from the optimistic side as my bosses frame it: we're all moving up in the ranks! Well, my team already has a PM. She's already trying to vibe code her PRDs. Thankfully, right now, they suck. But they won't forever because all of us engineers are being forced to train our replacement. There can only be so many strategists, project managers, designers, etc. This isn't simply the technology changing and us adapting, this is a massive reduction in workforce requirements ACROSS industries. Hype bros thinking we'll all have jobs once the AI is good enough to replace us are simply hallucinating as bad as the frontier models of 2024.
There’s no demand for human dexterity anyway, and it’s not about money. I like doing intellectual work.
They are spending 1T to ruin the quality of work we do, also probably ruining our chances of securing a livelihood and then they want us to for some reason welcome it with open arms and openly say look how great this is and marvel at human obsolescence. Yeah no thanks.
Wowee, a post from the CTO of some AI slop company talking about how everyone getting replaced by the shiny LLM gods is a good thing, actually, how original!
There are two key elements in the current AI fearmongering:
1) People think they can predict the future
2) People think humans can't adapt
The number of disruptive "game changing" technologies throughout history is immense. There's no question some people get negatively affected, but humanity as a whole has prospered because of these disruptions.
Literally every time, including during the industrial revolution, people thought it was game over for society, good manners, the youth, the planet etc.
Keep cool and try to adapt to the inevitable change. The next generation will certainly do it and leave you behind if you stay in a corner denying the new reality.
Adapt, sure, but is my life better or worse after adapting? It's a fair question and a fair worry. The goal of most AI companies seems to be to replace some amount of labor with AI. Look how that worked out for our factory workers. They adapted by getting jobs at Walmart and Dollar General, or scraping by in the gig economy.
> They adapted by getting jobs at Walmart and Dollar General, or scraping by in the gig economy.
The tech crowd used to say that only the "dumb" ones did that. They said those who were "smart like us" were much more adaptable, learning to code and whatnot to find new, often even better careers. It will be interesting to see if the emperor has clothes.
No, and obviously neither can you, else you would have explained yourself. Best to not post while under the influence of silly emotions. That is what those tech people who called those manufacturing people "dumb" did and, like this, wasn't useful.
Silly emotions? I'm not the one gleefully celebrating the ruin of others (including families). You've chosen to 'other' a broad group of people with diverse experiences and backgrounds. "Those tech people" certainly include some people who called manufacturers 'dumb', but it also includes many (most?) who didn't, and even some who used to be "manufacturing people".
You say "who cares if they suffer? They're obviously all just bad people". The people who attacked blue collar workers said "who cares if they suffer? They're obviously all just dumb people (and thus bad)." So the bulk of people who you decry are not guilty of what you claim. But you are!
> You say "who cares if they suffer? They're obviously all just bad people".
This comment aside, I do not see that text anywhere, lest my own comments. Get back to us when you return to reality.
> The people who attacked blue collar workers said "who cares if they suffer? They're obviously all just dumb people (and thus bad)."
Whether or not calling them "dumb" was justified, there no implication of "who cares if they suffer". In fact, in the context I remember, those comments were made around trying to figure out how to help those who were deemed "dumb".
Why is it whenever people try to downplay the efficacy of LLM’s evangelists go “there’s literally never been anything like this,” but the moment people critique them or try to weigh their impact on society the response is “well it’s nothing we haven’t seen before”?
The amount of money flowing and the sheer scale of all of this stuff is without precedent. We are seeing governments from the local to federal level literally shift their economies because of all the speculation. We are clearly experiencing a moment of sorts, and anybody who is worried about what the fallout will be is just handwaved away as ignorant fear mongers. Unless you are the head of these companies doing it in a disingenuous way to pump up your valuation, then suddenly it’s legitimate - but not so legitimate that we should do anything besides let the US government build a moat for them!
> The amount of money flowing and the sheer scale of all of this stuff is without precedent. We are seeing governments from the local to federal level literally shift their economies because of all the speculation.
Yes, the absolute numbers might now read TRILLION instead of BILLION or MILLION, but the concept of "Technology so impactful governments shift their economies" is in no way new, and has already happened a bunch of times before in humanity's timeline.
I guess that's why you're seeing some of us less stressed out about it, as again, scale is different, because scale today is different in everything, but relatively, we're just repeating another of those "Technology changes everything" milestones that we've done before.
We have never seen governments essentially leverage their economies so heavily, so quickly, for a speculative technology. It is literally propping up a significant portion of the US economy right now.
The US canal and railroad boom in the 1800s definitely had exactly the same thing happening, just compared to our scale today where everything is "more" and "bigger", it looks comparatively small. But in their world, back then, it was basically the same thing, just with canals and railroads. Indiana almost financially ruined itself because of their "infrastructure program". The government got involved heavily with the "1862 Pacific Railway Act" as well.
And yes, easy to say now that "Duh, that's essentially infrastructure" but at the time, that wasn't as obvious to everyone, as and railroad were newfangled technology, especially deployed like that, on a scale never seen before.
> definitely had exactly the same thing happening, just compared to our scale today where everything is "more" and "bigger",
And faster. More, bigger, faster. Which substantively changes the situation the same way a musket stops venn diagraming as a “firearm” at some point when compared to a 50-round automatic rifle.
You're mixing a bunch of different "people" in the same bag. I'm not the one saying "there's never been anything like this before".
The investment is massive, but not bigger than other projects by an order of magnitude. Major wars were bigger than AI. The dot com bubble was 2-3T dollars.
There may not be so much in the way of soldiers in the field guns blazing, but we are very much seeing a 'cold war' emerging with AI at the centre of it. Call it speculative, but the incentives for nations to spend are essentially the same in cold and hot wars.
> Many of us, deep down, are driven more by the joy of making things, solving problems, and seeing people use what we built than by the act of coding itself.
I LOVE playing "product designer" with LLM. The thing I normally hate and never do, write and update "design docs" being my only lever is awesome. I do already hate how it seems to turn development into a much less social activity. It's great that we don't need other people and a lot of time for some things now, but it's also not.
But more importantly, because I also like solving problems and seeing people thrive, I can't overlook that it's all based on stuff taken without consent, how artists were made "obsolete" by what plasters me with slop every day (already, and we didn't even get started). And they're not even going "okay, now that we proved it can work with stolen stuff, let's do it proper". No, this is "the new world now", just keep pushing and squeezing and extract more "value". So wtf? Well, I like solving problems, I love games! One cool game is "catch the thief", another is "call out the sociopath". So let's?
But we can't, because we're not building this after deliberation and discussion, we're being herded into it and then make rationalizations for foregone conclusions made in our absence against our interests. Building systems nobody understands that keep making more systems nobody even tries to understand is not some hyper-efficient future, it's no future at all, and nobody with power (however wishy-washy that phrase is) wants to live under any of that, you can be assured.
So this all seems more like an airlock to me, an abattoir, not a spaceship that just happens to be kinda dirty and stinky because the focus is on so much grander things. How insincere and stupid so much of it is just can't be overlooked, and very much unlike other technological advancements where the results speak for themselves, where the results (or the improved process to get them), pull something, make it inevitably seep everywhere it is useful. There is so much pushing and hyping, and so precious little good stuff to make for the really bad stuff, all the slop that is produced and clogs up inter-human channels.
This is not even progress, this is just subversion, that's my thesis. It's deliberately driven for the easily predictable bad effects it will have.
Humans ourselves are a technology. We have incredibly flexible brains and dexterous hands. Up to now, we could always do things that robots couldn't. That was where our job security came from.
But what happens when a robot can do anything a human can? When a cheap machine can do anything you can do but faster and better, what makes you different from a horse in the 1920s?
This is what the article misses and what people have trouble accepting. Yea, this is a technological revolution, but unlike the Industrial Revolution, you can't just upskill your way out of it.
> But what happens when a robot can do anything a human can? When a cheap machine can do anything you can do but faster and better, what makes you different from a horse in the 1920s?
I've already started to feel this.
One thing that made us special was our intelligence. We were, and were always going to be (unless some alien form of life arrived), the most intelligent things on earth. Is that still true today?
We should start with the assumption that humans are valuable and deserve to live with dignity. Any philosophy that doesn’t have that at its center is broken.
We should, but to date anyone that wanted to be treated as though they were valuable and deserved dignity needed some way to apply pressure to ensure that.
I'm not optimistic that this will change in the future -- but, I am watching automation erode the amount of pressure that humans can apply.
Yes but same should have been true of animals and humans as a whole did not afford dignity or value to animals. Maybe we will get a taste of our own medicine this time.
Depends on who you ask. Some people might put it at opposable thumbs. Others might ask how many legs one has. Or language, or ability to count (crows can count up to 30). Memory (elephants have memories that go back over a decade).
Very often the justification for why these things aren't terribly disruptive is that the very best of a certain trade or field will still have a useful skillset. Like the best graphic designers in the world are still more creative and skilled than AI, and the best writers are still better, the best programmers are still better, just that they're all more expensive and slower.
But we're not taking into account how it affects the mediocre or average people in these fields. Roughly half of all of these people are below average. Maybe if you're the best you'll always have a job, but what about the guy who legitimately struggled to finish high school? The person who genuinely doesn't have the capacity to earn much more than minimum wage?
We're currently at the part of the story where chess AI can beat amateurs and is rapidly improving. Kasparov's not worried, he's an expert. He's more creative and skilled than an AI.
Why would these machines stop improving at average this time?
I didn't say they weren't improving. My point is that if our justification is comparing the top 10% of performers to AI to justify human performance, that speaks poorly for the replacement of the bottom 10% of humans.
I weep for people whose scientistic reductionism has led to such devaluation of humanity. Robots don’t do anything and don’t think anything. Humans do things with and through them. What makes us special is that we exist and love one another.
The economy long moved past the things we can do, hence "bullshit jobs". Our job security has come from our willingness to "look the other way", allowing the job creators to live out lives the rest of us can only imagine. If they didn't placate our insatiable desire to work, that wouldn't last long.
The western world, the US in particular, already has outrageous abundance.
And yet problems of despair, poverty and marginalization continue to exist.
5-10% of people experience some level of food insecurity. Is that really because of a lack of resources??
It seems to me like we have a resource allocation problem, not a resource generation problem.
This is not an argument for some form a socialism. But instead an argument that progress in the field of social organization would be more productive than progress in the field of technological disruption.
"My family" - a nice anecdotal frame. No worries here. After all, labor savers for production has evolved for millennia. And you should only worry about the impact to yourself, and be glad your lot is better than your ancestors'.
But our society today has very extensive systems with very few deciders, who are largely enmeshed in social and legal constraints to ensure our reliance is safe and effective (at least in the large/long run).
The concern is about that constraint system being displaced - minimally with decider over-reliance on AI (still subject to guardrails?), or maximally with AI itself (no internal guardrails). Slop (in software, medical services, and military targeting) is just the symptom.
That replacement/shift has already been a corrosive side-effect of capitalism, as elites slowly cede loyalty and professional standards to competitive pressure, but even now most constraints allow replacement of bad actors (if demonstrable). AI reliance just makes that shift irreversible. People no longer have incentives or opportunities to spend the decades required to lead effectively, so organizations could never revert the heartless efficient machine elites back to the available incompetent and oblivious humans.
Maybe the result is just wealth transfer, or health, but I'd expect the high-value use to be in a military where very few people can combat many (as is happening in Ukraine now), making the military a preferred option for anyone with resources.
The distribution of (and checks on) responsibility - up, down, across companies and political systems - was intended to constrain agents to organizational goals, but it effectively kills any internal change by requiring near-unanimity and overcoming people's dependence on their jobs. So even if a super-majority of people agreed on some negative projection of AI usage AND had the resources to put work and investments at risk, it still wouldn't be enough to change this trajectory.
More AI deepthroating with the usual psychological-care veneer.
No, the difference this time is that the goal is to replace all coding, and perhaps all office work. Two obvious problems:
1. All of Office Labor has nowhere to go to—how to translate thirty years of experience staring at a screen into something else? And will there even be other jobs for them?
2. No labor leverage since there is no labor in that sector and the capitalists just capture all of the value for free (with an initial investment in CAPEX...)
How does this compare to horses and cars? You still have to get around.
Granted these are the maximalist AI Cretin goals and I’m also influenced by their relentless propaganda campaign.
> Many of us, deep down, are driven more by the joy of making things, solving problems, and seeing people use what we built than by the act of coding itself. After all, many of us decided we wanted to be developers before learning to code, just to create video games or websites. The desire came before the code.
News to me. As long as I was in this wider culture, coders were sneering at people who were in it “for the money”[1] and not for the intrinsic pleasure of very specifically typing out and testing code. And at ideas people. But as soon as these programs hit the mainstream, oh the most vocal ones just never cared about coding itself and they were in fact passionate about all of those business value-adds they were adding.
And managers? Clueless people herders. But what are clankers comparing GenAI coding to? Management. And this is enjoyable to you?
People will yell Goomba Fallacy. Of course maybe that whole old guard was just displaced.
in about 15-20 years when all office jobs are gone and we're only talking to machines there is going to be "ai alignment and safety specialists" doing the same jobs in smaller offices for less pay than customer support does today. they'll be fixing "misaligned" models and penalizing citizens for deviating (see credit scores, penal system today) all from the comfort of their home.
they'll be more afraid of terrorism, black people and muslims than ever. and lets not get started about socialists!
however you spin this it is going to deteriorate your average job opportunity at least in the us. the us will be where China was 50 years ago: worse workers rights, worse political participation, more low income jobs working for "miltech superintelligence corp".
what hegseth does to your military, amazon-ifying it is what is also happening within your political system stripping you of your civil rights.
at least china had an economical goal furthering the nation state. maga has... what? sucking up to elons, zucks, bezos and other super rich weirdos in the hope of resurrecting the big american empire again. it's devastatingly stupid, really.
Aside the fact the author sells ai services, training, whatever i can not understand the view how knowledge workers will be replaced while every employee is now forced and expected to work 10-fold and hiring is halted because shortsighted deciders and ... money.
The effect on the market is pure GREED and wishful thinking as none of the people in charge of HR or decision making have any idea how this technology actually works.
Also - why is everyone comparing this to the industrial revolution? And not f.e. invention of vaccines. No it HAS to be the big big story or you cannot make money with it.
All while the actual results are well, great. Helpful, sure. But so are vaccines!
Don't buy into this bs, any service or course. d2l.ai is free and wont be stupidly useless next month like whatever hype we sell this week.
As the doom stories lose novelty now we get the feel good "my grandpa" stories.
Thank you for your comment. I didn't intend to tell a feel good story. I'm just a seed startup founder, I'm not DHH! I'm just sharing a personal story that helps me navigating into the storm. There is no message here, I guess I hadn't been clear enough.
Author here! Thanks for all the comments. This post isn't meant as a lesson or a judgment, something to tell "just shut up and adapt like my ancestor did". The situation is hard, and I wasn't trying to dismiss anyone's anxiety. No one wants to hear "you'll be fine" when something bad just happened.
It's mostly a personal story. You can see it as a kind of tribute to a great-great-grandfather I never met, partly because I'd never heard a story like it before. It's banal but still a big adventure in the family. I find it fascinating that he grew up in a completely different world from mine and still felt some of the same doubts and anxieties I do today. And I'm really proud of him, because his "pivot" let my family stay in the village instead of maybe leaving to work in factories in the city. My father loved his job, and in a way that's thanks to him.
I appreciate your wholesome message, but I feel a bit is overlooked.
Horses and engines are in a tangible, corporeal domain. I almost feel like pivoting could be easier with that: the tools are still the hands. Also, pivoting might be easier especially when no one is in that space of repairing cars in a small village at the time. Low competition. I also think time allowed to pivot is much higher: automobiles were expensive and time to adoption is higher, I wonder how long it took to be able to make a living?
Data, numbers, programming, this is in a ephemeral domain. Competition is high, there are a lot of developers. Time to adoption is extremely fast. In the automobile era, someone may have had five to ten years to pivot. Now? Maybe a year?
Is it about making things? Maybe, but pivoting and making a career of it is not a straight path, and sometimes people get lucky in choosing the path.
Interesting historical read:
https://historycollection.com/10-industries-either-killed-cr...
>> Many blacksmiths added the mechanical skills needed to service and repair automobiles, particularly in smaller towns and rural areas. In short order, some of the more commonly replaced automobile parts, such as tires, were being stocked by blacksmith’s, others which were necessary to a vehicle’s repair were ordered by telegraph or telephone for delivery to the blacksmith’s shop. Blacksmiths and service stations soon had competition from the automobile parts store, another new specialty which blossomed along the popularity of traveling by car.
Comparing programmers to farriers is definitely strange on the surface. However I think it's quite unfair to reject the authors premise due to the ephemeral nature of some software. And I think it's downright arrogant to assume moving from horses to cars wasn't seriously challenging for his ancestor. But yeah, I would love to chat with the old man about what that transition looked like.
Also, AI adoption might look like it's moving fast for individuals, however I see many companies struggling to actually utilise it effectively. Business processes are slow moving political machines. Sure a large part of the software industry is ephemeral, indirectly providing capital value. However many jobs are still very tangible; embedded, backend web systems, networking etc etc. Instability in these systems absolutely results in tangible loss of value.
It will absolutely take decades before we figure out 'correct' AI usage at an industry scale. Heck, even frontier labs are still undecided about best practices.
P.S. ty for the article link, will read!
> Comparing programmers to farriers is definitely strange on the surface.
Just for another datapoint: two of my great-great-grandfathers in two family branches, in the same small village in France, were respectively farrier and black-smith, while I'm a software engineer. One son of the black-smith continued the trade as a side activity of being a plumber. So I had to think about how my career choice (and those of my siblings, uncles, cousins) relates to theirs.
My own perspective on those trades is that transforming metal was the top engineering scale in the countryside in their era. In those times (XIXth century, early XXth), learning the job required to go work with different masters around France ("tour de France" had another meaning before cycling). While it was still manual work, it was highly skilled job which required both technical skills and creativity, which are also key skills for software engineering.
Another thing is that, unlike AI, automobiles were (somewhat kinda) going to "replace horses".
The AI situation is similar to someone saying "see, we're gonna keep the horses exactly as they are, but replace farriers, trainers, grooms, stable hands, saddlers, vets, and coachmen with absolutely nobody other than some manager type".
The AI industry can't really say they're gonna replace SaaS, because SaaS is the lifeline of their business.
So they just claim they're gonna replace several classes of professionals.
> In the automobile era, someone may have had five to ten years to pivot. Now? Maybe a year?
ChatGPT was released in 2023 and we're still not out of the job for the most part. It doesn't feel slow but we might have these five to ten years.
Yes, but do any of us have any idea where to go?
The total market addressed by current-shaped LLMs is not "software development" - it's approximately all of white-collar jobs. And with recent advancements in robotics, blue-collar work isn't really much safer. Hardware is getting better, and the one thing that was stopping robotics from entering our own workplaces and living spaces, the ability to make decisions on the fly in heterogenous, unpredictable environment - the judgement - is literally what LLMs themselves are embodying.
So I agree, we all likely still have 3-5 years to figure this out - but also no indication where to go, or if there even exists a direction to go in.
art and music for some, but personally I'm most interested in seeing where science goes. Reproducibility of experiments is gonna be first. How many experiments don't reproduce, or have unwritten biases in the papers. Room temperature semiconductors. Fusion seems actually a decade away, finally. Research into better treatments for diseases, how to help people be smarter, live their best life. There's plenty more to do now that making CRUD apps is so much easier. Maybe AI will become creative enough to not need humans for that, but it'll still need someone to drive it and choose which direction it goes.
> There's plenty more to do now that making CRUD apps
Throwback to HN anno 2021 everyone was complaining about their bread and butter was creating CRUD apps.
We didn't expect we'll automate high thinking faster than menial labor.
Photonics research, to make faster AI chips :)
Art and music? Ok for live performance. But recorded performance is directly in competion with genAI for revenue streams, with content used without permission for more genAI training.
AI will do all of what you listed better, faster and cheaper by maybe Opus 9 or so.
It can't though. You still need to run real trials on living humans, living their lives and taking experimental drugs/procedures/interventions and running statistics on them.
AI can do everything you described except the "living out a human life" part - but it can iterate on ever better simulations of that, or worse, it may reward-hack and... well, I'd rather not be the human in that story.
Do you really think humans are going to be the users? Sure for tomorrow and next month, but the clock is ticking. I got mad at our internal CS peeps the other day. They were demanding an internal tool. I was like we went through hell this summer replacing the front end so it's agent usable, just get Claude to do that. They were like oh...
November 2022
See also:
Horses: AI progress is steady. Human equivalence is sudden (December 2025)
https://news.ycombinator.com/item?id=46199723
Sounds amazing. How on earth did he go from shoeing horses to learning about the internal combustion engine? Seems quite the leap that not everyone would be capable of.
He just vibe-repaired cars!
Joke aside, it was not shoeing horses but also working on the carts, etc. Probably bikes too. So at least he reused some mechanical skills, and probably tools, he already had. Only the engine was really new.
And many people living in small villages back then had basic knowledge in many disciplines, including electricity. When you had a farm for instance, you needed to be able to repair yourself anything.
This happened to my grandfather (blacksmith to working on cars and motorcycles), who was illiterate. From what I gather, it was a lifetime of trial-and-error-and-keep-what-works.
And he was probably very inquisitive too. Great story.
Early cars were called horseless carriages for a reason.
Car bodies were carriages for few first decades of car history. So if you can fix and maintain horse carriage - you can maintain car carriage.
Engines were simple too and designed to be maintained in a field environment with few tools. So it’s reasonable that this skill was learned, especially by mechanically inclined people.
Early automobiles were very simple machines. I you knew how to repair a carriage or a wagon, you were pretty close to being able to work on an early automobile. You should check them out.
Conversely, carriages (and other pre-automobile machines) were also surprisingly complicated. Steering mechanisms, leaf springs, brakes, and a variety of linkages including the horse harnesses themselves (which had to account for the animal's gait).
Honestly I think people forget this fact. You don't have to go that far back to find vehicles that you absolutely could do anything to yourself with basic tools.
We're fixing up a 1987 Jeep Wrangler for our son. I could strip it down to the frame with 3 or 4 tools. It's kind of amazing how simple cars were, and that was in the late 80's.
My first car was a 1986 Grand Marquis and I spent every other weekend of high school fixing something on it. What an incredible waste of time: I worked a part time job so that I could afford to keep a car running so that I could drive between school, home, and my part time job instead of riding the bus.
But it was easy to work on. My dad (not a mechanic) usually had a strong hunch about what the problem was. Lights went dim while you were driving and five minutes later it sputtered out? Alternator is dead. Flip through the manual to find a diagram of where all the bolts are, take it off, run to Auto Zone, get another one, bolt it on. By the end it was usually dark and I was finishing the job with a shop light hung under the hood.
I drove a series of beaters from the 80s and 90s and did all my own repairs until my mid 20s. Alternators, water pumps, break pads, break lines, ignition solenoids. All of it is just parts that you unbolt and bolt a new one on. When I got my first modern car (a 2008!) one look under the hood made me realize I had joined the yuppie class and would not be changing my own spark plugs anymore.
I always did my own maintenance including changing spark plugs. Then I bought a Subaru. That weird 4 cylinder has plugs facing slightly down that you either have to cut the wheel well cowl or UNMOUNT AND JACK THE MOTOR OUT OF THE ENGINE BAY to change.
My solution was to use like 11 wobble extensions and pray that I didn't drop anything. 8 hours later I changed 4 plugs and understood why it cost $250 at the dealer to have them to it. Then I gladly started paying that money.
Yeah as long as you ignore the engine (ECU/PCM/TCUs) in cars in the 80s, they're quite simple. Go back to the early 70s-before and it's mostly all analog or trivial electronics.
> Yeah as long as you ignore the engine (ECU/PCM/TCUs) in cars in the 80s, they're quite simple.
Alec of Technology Connections just posted a video on this:
* https://www.youtube.com/watch?v=0KoeY54M5VA
This is a great video, thank you.
As an aside, because of this Technology Connections video, particularly from: https://youtu.be/0KoeY54M5VA?t=746
I now better understand the voir dire scene from My Cousin Vinny: https://youtu.be/1QrTzTe90QA?t=200
I had always loved that scene (as well as https://youtu.be/CFdJza0AbeA ), but I now love it a little bit more. :-)
This was the video that earned my sub. Even though it covers stuff I mostly already knew, it's done in such a way that that it breaks down the whys that may not be apparent to people without a background in this stuff.
This and the video on hybrids, are, imo important works, because they break things down in a way that disarms tribalism, and explain how it is that we got to where we are. There are very real stakes in ownership, right to repair, and freedom of movement that rest on people having a good baseline understanding of this stuff.
Even as late as the late 90s, probably into the mid 2000s for manuals. The complexity of vehicles from that era is mostly in understanding the role of the sensors (or that they even exist in the first place). Once you understand that, they go back to being pretty straightforward. I guess the exception here is probably immobilizers.
The complexity of modern vehicles comes down to software coupling between the systems, which (to my knowledge) started with automatic transmissions, and then spread to everything else. In very modern vehicles, even the sensors register with the ECU and communicate on encrypted channels.
The reason for these measures is ostensibly security and quality, but automakers don't seem to mind the fact that it also makes many repairs more difficult or impossible.
Shoeing horses properly is actually no joke and cars back then were laughably simple. I don’t see how any of that could feel relatable to people in software development now, when almost all jobs that require real smarts and competency require years of study and experience.
I don't think the story is meant to be taken as literally as I'm reading in this comment.
Many people in life are underutilized and ready to step up to challenges if given the opportunity.
+1000. The best part of being a manager is when you can help a person achieve their potential.
In the late 1800s there was a cottage industry of mechanics and ironmongers selling and repairing small steam and kerosene engines used by rural mills and farms. There were horse drawn steam threshing machines. Bicycle repair shops sprang up. The Wright brothers got into the bicycle business before horseless carriages. There were many intermediate tech skills that led up to working on motorcars.
Out of curiosity, what was the timeline ? Cause I'm afraid that, when you mentioned "I don't know what the other farriers" did, the answer may be "they were sent to the front and died by the 1000s".
Which starts to be my fear for the real "solution" to job displacement by AIs : people will just be sent to some far away conflict where they will transition to a career in "daffodil fertilization" :/
I mean, the wars are also getting automated now. Sooo, no need to worry?
My father worked for the railroad as a blacksmith but he apprenticed to an old blacksmith who was a farrier before. There are still farriers to this day.
Some of us will need to move on to other things but there will probably still be a lot of farrier type jobs around.
I appreciated reading the Barrois territoire quoted :)
Don’t worry about it. You’re just playing your part. The trite comparison to past successful innovations. The paragraph about the AI Revolution. The AI Inevitability Soothsaying (“Be willing to let go of the how to hold on to the why”).
The perfect HN frontpage material.
Of course I wrote the two preceding paragraphs before reading your webpage: “I am Co-founder and CTO of Tandem, an AI Assistant for SaaS apps.” Just another entrepenur with a psychological spin.
[dead]
Shallow dismissal.
Please consider that this family story likely motivates OP's worldview and projects, rather than the other way around.
The flagging here is better than in a Steinberger thread. Unreal and congrats.
Steinberger as in Emil Steinberger?? I don't get it. what's the reference?
There is a quote from a CGP Grey video from a bit over a decade ago, "There isn't a rule of economics that says better technology makes more, better jobs for horses. It sounds shockingly dumb to even say that out loud, but swap horses for humans and suddenly people think it sounds about right." [1]
A couple hundred years ago, something like 70% of the population worked in agriculture. Technology replaced almost all of these jobs.
So far, we have reimagined old professions and created brand new ones. Is this ability of ours limitless?
[1] https://m.youtube.com/watch?v=CMFj75kBQlU
This is one step away from walking into the same mistake all these discussions do, including replies to your comment.
Yes, we can probably invent new jobs and new career fields.
No, it does fuck all for people who lose their current jobs mid-way through their career or later, and watch themselves and their families be kicked down 2-3 rungs on the socioeconomic ladder, possibly throwing them into generational poverty.
Or, in short: yes, there will be new jobs. But not for you, and not for your children.
For the society and human civilizations, these techological revolutions were always a net positive down the line. For individuals caught in them, they rarely were.
That's why the Luddites were destroying looms. Not to fight the progress of technology, but to save their own families from being thrown into poverty.
Timeframe is also different, agriculture changed over generations and the kids that would not take over the farm had access to education and could move to cities. Now we may see a similar job destruction over a decade without alternatives available for much more people.
> had access to education and could move to cities
Or their rural economy collapsed, everyone was hungry and unemployed, and they moved somewhere new out of necessity and basically started over with nothing, living in squalid conditions and working very unsafe, unpleasant, and exploitative jobs.
Yeah. But now we have nowhere to move, either.
I think the difference is the Luddites weren't happily designing and building innovative new looms for 50 years and now suddenly surprised-Pikachu that the automated-loom-maker has come for them.
What did you think "disruption" meant? Vibes, papers, new Javascript frameworks?
Yes, because people are consumers, horses are not.
Jobs are a mean to make people consumers, not the only mean, but a significant one.
The ability to create more economic activity require both sides to catch up, so we will find a way to do so.
The amount of consumption is largely dependent on income level. There's a recent study from Moody's that says that in the US, the top 10% income bracket now accounts for roughly half of consumer spending. As wealth inequality continues to worsen, it makes more and more sense to solely target high earners because you make the same or more as selling higher amounts of lower priced product but with less overhead.
This doesn't make sense at all. It is a bad statistical analysis of consumer spending. [1] A few things from the thread on why by Levy on why it feels wrong .
First, why does it feel wrong?
1. the top 10% get about 50% of pre-tax income
2. they get 30 to 37% of disposable income after tax&transfers (depending on source)
3. and we know the rich have a higher savings rate, so consume a lesser share of their income (US, China)
[1] https://x.com/LevyAntoine/status/1985127826207772920
Could the original chart be including non-cash consumption? (Private jet flights paid for by company, stocks, bonds, etc)
> the top 10% get about 50% of pre-tax income
May be not top 10% but top 0.1% are good at avoiding tax using various loopholes. Like borrowing against stocks (and other appreciating assets) till death to avoid paying CGT and debt is then repaid by heirs which do not pay CGT AFAIK. So they can spend more than income stats would suggest.
> As wealth inequality continues to worsen, it makes more and more sense to solely target high earners because you make the same or more as selling higher amounts of lower priced product but with less overhead.
This is exactly why everything like Disney has moved upmarket, with pricing rising far above inflation.
It works right up until some form of revolution, either politically or violently. We are far from anything like that but I'd rather not get close to the line. I think I'm a minority view though, despite being in the top 10% myself.
It is also shocking how little my life compares to the rich people I see. I don't own two houses or a yacht. We don't take yearly vacations to a chateau in France. The gap between the top 10% and the top 1% is insane. The gap between the 1% and the 0.1% is even larger still.
But hey... not like the rich getting richer and saying "screw you" to everyone else has ever led to negative consequences right? Nothing to see here, business as usual.
Did anyone ever stop Marcus Licinius Crassus from destroying the Roman Republic? (patron of Julius Caesar and Romes richest man ever).
Caesar was a populist reformer. The Roman Republic was long destroyed at that point, due primarily to concentration of wealth and replacement of the plebians with slaves.
First Triumvirate.
Crassus amasses so much wealth because Caesar helped him.
And Crassus was the rich man who was monopolizing all of Rome with his slave firefighters + real estate scheme. When the going got tough, they seekwd power from the rich guy, no one was stopping Crassus.
I think you have the wrong end of the stick (trying to map it onto contemporary politics?).
Caesar was killed by the same people (hardline conservative ultra-wealthy 'Optimate' families/factions) who also brutally murdered the tribune of the plebs (person appointed to beg the ultra-wealthy for favorable legislation) and his supporters not once but twice. He was killed for the same reason too - land redistribution, limits on slave labor. That was called 'tyranny' by the elites.
Crassus was wealthy before Caesar and died before Caesar took power. This kind of politicking was just a necessary evil in the Roman Republic - which always was an oligarch-controlled plutocracy and never a true democracy as we know it today (e.g. Athens). It was 'pay to play', sponsorship of the ultra-wealthy was not optional.
[dead]
The Parthians sure did :)
If by Disney you mean tickets to Disneyland and similar theme parks, then pricing rising above inflation actually reflects wealth inequality getting smaller, not bigger. A lot more people in the world than in decades past now have enough wealth to consider traveling to one of the small number of Disney theme parks in the world and spending time there, and there's only so many people who can fit in the park at once. The Disney corporation could (and has) built more parks around the world, but maybe you care about Disneyland specifically, and there's only one of those today just like in the 1950s when it was built.
What are you talking about? We can actually track wealth inequality and it has not lowered at all within the last 50 years lol.
https://en.wikipedia.org/wiki/Wealth_inequality_in_the_Unite...
https://time.com/5888024/50-trillion-income-inequality-ameri...
There's a lot of people who live outside the united states.
> There's a lot of people who live outside the united states.
True... but not really relevant here.
Wealth inequality is a global problem.
https://www.theguardian.com/world/2026/sep/24/un-global-ineq...
> This is exactly why everything like Disney has moved upmarket, with pricing rising far above inflation.
Also the logic of what Broadcom/Avago did with VMware: only chase the whales.
The issue is that there isn't economic demand from that audience for approximately 90% of companies operating today. Maybe Disney can succeed by going upmarket, but can all the animation/filmmaking/TV/theme park companies they compete with?
The number of millionaires/billionaires/etc feels low enough that there won't be demand for dozens of companies in every possible industry. The ones that have the biggest fanbase/cultural footprint maybe stick around, but the majority would presumably just go under.
> This is exactly why everything like Disney has moved upmarket, with pricing rising far above inflation.
Yeah, and not just individual destinations but entire destination cities.
See, for example, the radical reshaping of Las Vegas.
Horses absolutely are consumers. They eat, they drink. They just can't manipulate their environment to get more food or drink.
The human system relies on us doing stuff for each other, mediated by "money". We get "paid" by some other human for doing stuff for them. "Stealing" is discouraged. "Being given" is rare. "Money" originates from fiat, goes where it goes.
Humans unable to do for others don't get "paid", get starved/sick etc, experience pain, cause trouble and disappear eventually. When doing stuff for people is not done by people, the pay cycle breaks. When machines do a substantial fraction of economic activity (intangibles like decision-making, writing, answering the phone, etc. as well as transport), there is no need for very many people. Absent the need, there will be no support. The farms won't give their food away, they will arm themselves against the hungry.
There will be some human work. So long as machines cannot equal human touch and experience combined, there will be medicine. There will always be sex work. But who affords it?
Possibly racial disparities will diminish. A machine might actually give equal medical care to, or equally arrest, the Black and white person. Or not, depends on the training. Look who's developing the AIs.
Economists thought that economic development means democratic development. They were wrong, see China. They thought that economic liberalization means good governance. Wrong, see Russia. Economists are right about things until they are wrong, and we can all run off the cliff with them.
When the population drops (war, starvation, refusal to bear children) to the level needed to run the machines, the system will stabilize. Pretty much the horse scenario.
Yes, we've seen that already without necessarily a technological singularity. You accurately describe the great depression in the 1920s and several similar events prior to that.
But what's consistent with those events is that they end and economic development is resumed under different circumstances. I don't see any reason that the current era will be different or that current problem will be unfixable. People used to get paid to build national parks. What would be our current iteration of national parks?
This transition will be a lot slower than the shocks that produced depressions and recessions. A couple of good wars will thin the herd, which is how populations have been weeded out in earlier times.
Public works projects need funding and purpose. I'm not optimistic about what purposes the make-work funding gets applied to these days. Public beauty, cleanliness, health are not in the priorities.
Huh? The Great Depression was a cyclical event. So public works projects only needed to last until the market recovered. Technological revolutions are structural changes. There is no “national parks” equivalent program which we were willing to run to sustain the horse population at 1900 levels.
Everyone wants seniors, nobody wants to train juniors.
I imagine this is gonna continue to be true, except that "everyone wants consumers, nobody wants to employ people".
Does this result in millions of micro-businesses? All on the verge on bankruptcy, all continuously going out of business as new ones are invested in by VCs hoping to make a quick buck
More likely it results in a state of permanently doing gig jobs as any temporary work you can find is just a front to train a lower quality, lower cost automation to replace yourself.
This sounds like you are deriving an is from an ought.
Is it not conceivable that we won't find a way to do so, and that consumption and economic activity will ultimately decline?
That is one possibility but it goes against the interest of all income groups of the economy, including the ones on the very top. So a solution is in everyone's interest to be found, and nor is it a physical impossibility.
Until the ones at the top own a fully automated supply chain for their every need and luxury and all they need from everyone else is the ability to feel superior.
In this case, why doesn't the rungs of the economy below the rich just... run the way they have been? Serious question, I've been mulling over in my head what it means for a few people to own the majority of "the economy" and goods to be very cheap, but no one has any income.
What I always come back to is this: Why does it matter what the very richest do? Worst case the economy operates approximately as it does now (maybe with us knowledge workers knocked down a few rungs). But like, the rich can't eat _all_ the food.
I feel like the most likely outcome is that the quality of life of the bottom 90% is overall better than today. HN'ers quality of life is maybe worse than it is today.
They keep buying ten of the houses I want to live in, knocking them down, and building a single tasteless McMansion on the combined block, and then lobbying for building and zoning regulations that won't let me build another house.
Presumably if we're all unemployed or employed in jobs that have zero economic worth to them, they'll buy all of our houses, build a bunch of big grey boxes for us to live in like a human zoo, build even bigger and more tasteless McMansions for themselves, and pass legislation that makes it illegal to live near them or inconvenience them.
The quality of life for the bottom 90% of the world has been getting better because they were so desperately poor that they could rely on economic growth to help them. At some point they're going to be rich enough in commodities that they start looking to buy things like houses or start businesses, at which point they'll find that asset prices are rising faster than their own wealth because of growing inequality, and they will look at their big grey boxes full of shoebox apartments and start to become unhappy too.
I think if we jump instantly to a fully automated world, perhaps there's no problem, but in a normal transition from the actual world the issues are
- capital / limited goods
- allocation of labor
Things like land are a limited commodity. Concentration of ownership usually comes with rent extraction. Because you need to live somewhere, but available locations are finite (and supply can even be controlled, and is). This means that you can end up with a cartel scenario where increasingly normal peoples previously disposable income is tied up in an increasingly large % to those fixed costs of living. This is arguably well under way in some countries.
The second point is, the richest enjoy goods like personal yachts, private planes, luxury restaurants. If a lot of labor is increasingly allocated to producing highly inefficient luxuries, it's essentially wasted. You might have hundreds of businesses employing thousands of people or machines, but the productivity of their work is essentially all going into products consumed by a few people. I'd say a limited amount of this is a good thing, it's a normal part of the proliferation of choice enabled by capitalism. But taken too far you return to a historic economy where most peoples productive life is spent serving a tiny elite, while living in poor conditions themselves.
In our financial system, your income as an individual has to be significant enough to send a signal to produce the goods you want. If this ceases to be true for even a significant minority of people, you will find consumer choices and available quality decreasing for those people, which is a real problem. This is to a lesser extent happening in some segments of the economy.
Yes, though I'd say that rent extraction and labor being directed to the whims of a small number of people is essentially the same thing. And even under a step jump to full automation, the owners of fixed goods can still extract monopoly rents.
This is also largely the conclusion of classical economics (i.e. Adam Smith) and the argument was developed in great detail by Riccardo, George (who explicitly considered the no-labor-required case in 1800s), and many others.
> In this case, why doesn't the rungs of the economy below the rich just... run the way they have been?
Where do you live? The rich will own all housing.
Narcissism and psychopathy are why. The richest people stopped needing to care about money to fill their lives with endless luxuries billions of dollars ago. They enjoy being richer and more powerful than others.
The rich will buy all the scarce resources and charge the maximum possible price to extract all of the value you can provide. You will be unable to accumulate wealth. Dynamic pricing of fast food will extend to all food and even to e.g. plumbing fittings - you don't own a copper mine, so you can't go make your own. You will do hard and degrading work for a subsistence living despite the ample resources and technology required to improve your condition.
This is history, not a random hypothesis. Feudal serfs had the ways and means to bootstrap, but in fact lived worse than their pre-feudal ancestors.
> ability to feel superior.
I actually have a random thought that it is this innate human drive that causes society to be mostly employed despite massive differences in technology and productivity across geography and time.
What if the driver of employment is not efficient return of capital, but satisfaction of human mind?
Your statement makes it seem like we had no problem losing professions to automation. I'd argue this isn't true.
There are plenty of poeple who have inadequate work the world over because there isn't enough paid work to do, and most professions are servicing wants, not needs, so the demand is very elastic, with many people now multi talented and still not making ends meet.
I know it's been talked to death already, but the fact that the video is so wrong on self-driving cars is mind-blowing. We can rationalize post-fact as much as we want - still it is shocking that popular belief never expected language models before self driving.
Popular belief never took the regulatory obstacles to self-driving cars seriously. If LLMs required insurance to use, they also wouldn't be available in public yet.
That's a good point. I'll add that killing someone (even yourself) with a public chat bot is a drawn out affair. With a car it happens all the time.
There are enough jurisdictions in the world where if the car really was good enough, they'd let it go. Even in the US.
It took some time for cars to be available even in good weather in specific places, so I'm guessing no, the cars aren't good enough.
I’m not sure if you understand how insurance is priced.
I'm not sure I agree that it was ludicrously wrong. His thesis is "the question isn't whether self-driving cars will replace human drivers, but how quickly". Sure it's been slow going, but you can't walk down a street in San Francisco today without seeing a Waymo, which wasn't true a few years ago. We're still at the bottom of the S curve.
1. There are no rules of economics. Economists attempt to understand things that happen and use that understanding to predict future events, but frankly there’s not a lot of consensus.
2. Economics is a function of human behavior, not horse behavior. Economics is about coordinating human effort. Humans make economics, horses don’t. When economics change in a way that is bad for horses, the horses have no recourse. When economics change in a way that is bad for humans, the affected humans seek to change things.
...by dying under the treads of tanks.
Humans are not required in economics. Only inputs and outputs. That's it. It is a grand delusion we have that we meatsacks are needed.
Civics and society seem to still need people and that's what I'm worried about breaking down.
Indeed. The economy may in fact eventually close in on itself and become a fully automated system with no humans in a loop. "Disneyland with no children", as I believe Bostrom called it.
Assuming we are around for this, this raises the question: if the economy can function without us, what claim will we have on its outputs?
Isn’t technology defined by how things can be used by humans, I.e. tools? Horses can’t use smartphones, and a “smartphone” designed for a horse would look significantly different. Seems a like a nonsensical argument based on the definition of words. Technology is what humans can use to achieve tasks.
The fallacy here is that horses were never employees!
They were tools that people used to get things done.
They got replaced by better tools. Just like typewriters got replaced by computers and printers.
> They were tools that people used to get things done.
This is exactly the perspective many executives have of their employees.
> This is exactly the perspective many executives have of their employees
There's a phrase for it in fact, "Human Resources".
Though that sounds more like natural resources, like trees for a lumber company or paper mill.
Probably there were people in charge of Horse Resources when horses were the dominant mode of transportation.
They still are, wherever people work with animals at small scale. People get attached, and the boundary between "animal as my fellow (if dumber) friend" and "animal as my main capital investment" gets really blurry. This was the reality for most people until 20th century, and we still see echoes of that in old books and stories.
Like there are people in charge of cows to produce milk and/or meat.
This sounds like a clever argument, but the mindset of some executives really has no impact on how differently horses and humans are treated in society.
To spell it out: When horses are no longer needed for a task, they are sold or sent to slaughter.
I hope I don't have to explain how differently you and I are treated when we lose a job?
To spell it out:
People need jobs in order to pay for food, shelter, and healthcare.
Many executives feel zero moral obligation to provides jobs to people and instead treat that as a necessary evil that gets in the way of profit. As soon as that evil becomes unnecessary, they will happily lay off as many people as they can.
In aggregate, when enough executives do that, then people will not be able to find jobs.
When they can't find jobs, they are unable to pay for food, shelter, and healthcare.
Even if the executives care, they are pressured by the board.
Even if the board cares, they are pressured by the investors.
Even if the investors care, they are pressured by the pension fund administrators.
Even if the pension fund administrators care, that far removed, they care about their pensioners more.
You're right that hardly anyone hires people because they feel a moral obligation. They need work to be done and pays what it costs to hire people.
Similarly, hardly anyone takes a job out of moral obligation to the employer. They need money, and accepts the best job they can get.
We've been doing this for many centuries now, and empirically it has been wildly successful for pretty much everyone involved.
> We've been doing this for many centuries now, and empirically it has been wildly successful for pretty much everyone involved.
In aggregate, yes. But if you zoom in on the trajectory, you can see that it's peppered with wars and revolutions.
Sure. But if you zoom out again, you'll see that the eras before was even more peppered with wars, revolutions, and genocides.
Also famines and diseases we can now easily cure wiping out entire societies.
I don't disagree. Trajectory has overall been great for people. Just saying it's not uniformly great for everyone at all times, in fact whenn it regularly flies off the rails, restoring it exacts a great cost from individuals and societies.
Maybe we don't really disagree on much at all.
I'm glad we had this talk :)
It's not really "many" centuries. It's more like three or four and even that's not uniformly distributed. There are still many subsistence farmers in the world, and almost all of us were doing that for thousands of years.
Also, those past few centuries have involved the largest violent revolutions and wars in human history.
The idea of employment and money is a relatively modern conception. There is no physical law that states it must work indefinitely and efficiently for all participants.
> To spell it out: When horses are no longer needed for a task, they are sold or sent to slaughter.
> I hope I don't have to explain how differently you and I are treated when we lose a job?
About the same once you run out of money. At least here in the US, ymmv in countries that have sane social safety net policy.
https://www.scotusblog.com/2024/06/justices-uphold-laws-targ...
Well, we don't currently eat horse OR human meat, so there's that.
Not just the typewriters. Also the armies of secretaries that came with them.
Employees are just another way to get things done. Ripe for disruption!
Lots of doom in here.
If software engineering goes the way of the dodo, I suspect most other white collar jobs are also displaced (anything done entirely on the computer, really). A few years after that I suspect an overwhelming supply of physical labor.
It’s a house of cards at that point, sure - software engineering could be the first to go, but I don’t see any reason to try to retrain or pivot. As soon as I lose my job and jump to plumbing, the rest of the white collar jobs are going with me.
If we get to that point, I’m just going to assume that the rest of the world is also going to be in shambles.
Though, I’m more of a Jevon’s paradox believer. We’ve had so much work in the backlog and now we can finally take a stab at it, I’ve been busier than I ever have been. Until the AI can competently run our entire stack, I’m skeptical. If it gets there, I’ve got no reason to try swapping jobs, because everything will be busted in a short timeframe - no?
I’m also trying to keep my eyes on the big dogs. Once Google starts to hemorrhage engineers and knowledge workers that means that AI can meaningfully replace entire roles, but until then I’m just sitting tight and building stuff, business as usual I guess…
I feel for everyone else in this position, on edge for the last 3 years with no end in sight.
> If software engineering goes the way of the dodo, I suspect most other white collar jobs are also displaced (anything done entirely on the computer, really).
Strongly disagree with this. In contrast to most jobs, software engineering is unusually self-contained, with short feedback loops and immediate validation. You (or Claude) implements the feature, you run the tests, you fix the bugs.
Most white collar jobs are nothing like this: lawyers have to prepare briefs (AI can help here) but they have to file briefs, they have to wait weeks or months for rulings, they have to argue in court. Marketing execs might spend millions preparing a new campaign, launch it for the holiday season, and it has to be right the first time. Most white collar work involves judgement, anticipation, and ambiguity in a way that software doesn't.
Software lets you guess-and-check over and over again until it works.
> Most white collar work involves judgement, anticipation, and ambiguity in a way that software doesn't.
In my current position as a software engineer, this is MOST of the job.
I don't know what everyone else's role is like, but a software engineer that isn't dealing with those three items sounds more like a compiler.
"Most white collar jobs are nothing like this: [software engineers] have to [implement features] (AI can help here) but they have to [coordinate releases across teams], they have to wait weeks or months for [real-world usage to validate architectural decisions], they have to [defend technical decisions to stakeholders]. [Engineering teams] might spend millions preparing a new [platform migration], launch it for the holiday season, and it has to be right the first time. Most [software engineering] involves judgement, anticipation, and ambiguity in a way that [isolated code production] doesn't."
Sure, and to the extent that your job involves those things, it won't be taken over by AI as quickly. But this whole thread is full of software people screaming that the sky is falling and Claude can already do their entire job and they are the horses, not the farrier. So...which is it?
I think software development is a field unusually vulnerable to LLMs because there's lots of good examples for training data, and the work is easy to validate in short order according to somewhat-objective criteria.
Note that certain kinds of mathematical proofs also fit these criteria, and also have been an early showcase for AI abilities. Anything where the bot can get to a good result by just grinding on it for days is going to get taken over by AI quickly.
> You (or Claude) implements the feature, you run the tests, you fix the bugs.
...you then take it to the customer. And wait, and wait, and wait, and maybe eventually they'll get around to trying out what you created, and if you're lucky they'll even give you some feedback to loop back with to try again. They will only try your application a few times before giving up on you, so even if the feedback loop is immediate like you claim (its not; not even close), you aren't afforded many iterations. A software engineer's job is, of course, to get ahead of the customer using judgment, anticipation, and dealing ambiguity to minimize how many round trips are required to satisfy the customer.
If AI can replace the software engineer, why can't it replace all "white collar" work?
For the record, I don't think AI is likely to replace those parts of the work, at least not in the near term (15-20 years). But at many companies, the people interfacing with stakeholders or users — deciding what problems need solving and how to solve them — are product owners or project managers, not software engineers.
I think there will be a convergence where AI enables more POs to create software directly, and more SEs move up the value chain to work on product-level problem solving.
Product owners/managers typically act as the communication channel between customers and software engineers, but engineers are responsible for designing the software. It is they who have to use judgment, anticipation, and deal with ambiguity in order to find software customers want to use. The product owner/manager does not. They simply pass messages back and forth. Of course it is possible for one person to be both a product owner and a software engineer, but often the engineer is separated from having to play telephone tag with customers so that they can more efficiently focus on their primary task.
Perhaps you have confused software engineer with coder? LLMs are most definitely the final death knell for anyone still working as a coder, but that profession was on the way out for decades already. The earlier comment was clear about software engineer.
Yeah, most PO/PMs I've encountered know awfully little about the business processes, their constraints and complex interactions within their own product. They know quite a bit about priorities the various customers voiced, fancy stuff they'd like to demo on the next review or trade fair or customer relationship meeting, or new features they want to get customers interested in. But don't ask them how those features interact with others!
If you want to go into detail with that kind of stuff, your best bet is not the PO, but the lead engineer of the team. If the team was properly staffed, the lead engineer will have a solid knowledge of the business aspects of the product while at the same time being able to cross-reference that with the most important technical aspects.
LLMs will make more people into "lead engineers", even if all they're "leading" is the LLM agents and not other humans, as it was common before the LLM age. One of those will then emerge as the "lead lead engineer" who's going to lead the humans which in turn lead the LLMs. If it wasn't for the "engineer" in the meaning of "code monkey" dying out - which I also expect - we'd have to invent a new title, but I'm currently expecting this shift to go without that. It'll still leave some engineers who made themselves comfortable just churning out code that does whatever a ticket requests, without even understanding one iota of the business process behind the ticket, without a job, but that's the kind of "software engineer" who in my opinion shouldn't ever have taken up the profession anyway.
> Of course it is possible for one person to be both a product owner and a software engineer
This was the norm in every job I had from late 90s to around 2015.
Engineering Managers were responsible for both technical and business outcomes, and it worked fine.
The scrum solution of having from the business side acting as a literal "product owner" or "client representative" and giving feedback IMO was a much better direction that unfortunately companies have not followed.
[dead]
I don't think "software engineering" is a real thing. So-called software "engineers" do not practice engineering in the way that a mechanical engineer or a civil engineer does. Consider: How often do bridges collapse? How often do elevators plunge to their doom? How often does some critical piece of software utterly fail? [0]
> They simply pass messages back and forth.
So you claim. We're about to conduct a real-life experiment to see which is more valuable: passing messages back and forth to the computer, or passing messages back and forth to the customer.
[0] https://en.wikipedia.org/wiki/2024_CrowdStrike-related_IT_ou...
> I don't think "software engineering" is a real thing.
You don't think anyone designs software? It just magically appears?
> So-called software "engineers" do not practice engineering in the way that a mechanical engineer or a civil engineer does.
And? The dictionary defines engineer as someone who designs and builds things. It does not prescribe how the thing is designed.
> How often do bridges collapse?
Not often, but often enough to know that we still consider the designer to have been an engineer and not a project manager. I still have no idea where the idea that project managers do the design work came from.
> passing messages back and forth to the computer, or passing messages back and forth to the customer.
When AI can handle the judgment, anticipation, and deal with ambiguity neither are important. The customer will go straight to he AI. Hence why "white collar" jobs of all sorts are ultimately doomed. Unless you believe AI cannot reach that level, in which case software engineers will be fine, since that's the whole reason for their existence.
What we do see happening already is the decline of the project owner/manager role. More expensive money and technology increasing engineering efficiency has made it such that it is becoming hard to justify an assistant to bridge the telephone tag gap. That doesn't really affect engineers much, though. Before ZIRP allowed budgets for project managers/owners engineers also played that same role, so it is nothing unusual for them. Engineering getting secretarial assistants for a time, something historically reserved only for the upper brass, is but a curious aberration.
> You don't think anyone designs software? It just magically appears?
Do you actually write software for money?
Most commercial software is never designed, it evolves, over the course of repeated "agile" "sprints". Every two weeks, someone hacks together something that kinda-sorta works and then that is released.
Eventually (often many years later) the thing becomes so complex and unwieldy that it's impossible to add any more features, so someone gets angry, throws it away, and starts again growing a new monster.
> Do you actually write software for money?
No. I am paid to be involved in the software industry, but as an engineer, not as a coder. Writing software is not part of what I do. That is, to use your bridge analogy, for the construction crew.
> over the course of repeated "agile" "sprints".
It is true that not all engineers know what they are doing, are good at what they do, or bold enough to push back against out-to-lunch stakeholders.
> so someone gets angry, throws it away, and starts again growing a new monster.
I never understood that. Untangling the messes was always the fun part back when coding was still a job. But I don't really see any coders in industry anymore. Any last remaining ones I knew of have already been replaced by LLMs. LLMs will also happily untangle the messes, though, so still no reason to abandon code.
Boeing ships software that crashes planes, Microsoft Windows regresses so you can't move the start menu, medical devices come with hardcoded admin credentials.
I'm very very curious where all these mythical software engineers are working. Finance, maybe? ... or not https://www.bloomberg.com/opinion/articles/2025-03-03/citi-k...
>> One quirk of the program was that the amount field came pre-populated with 15 zeros, which the person inputting a transaction needed to delete, something that did not happen.
> One quirk? Fifteen zeros? The quirk in the software is that if you type “send $1” it says “okay sending one quadrillion dollars” and you have to backspace over all the zeros? Incredible interface design.
> I'm very very curious where all these mythical software engineers are working.
Boeing, Microsoft, and medical device manufacturers mentioned seem like pretty good candidates. Were you under the impression that the software designed at those places just magically appears?
As before, being an engineer does not imply being good at the job. It only implies designing things. The designation of Professional Engineer institutes a vetting process to try and weed out those who aren't good at the job, which may be the source of your confusion, but PE is a designation, not a job. Not all PEs even work as engineers, although certainly many do.
Ok so why is all their software such garbage, then? You can't gaslight me, I work in this industry. I see the absolute nonsense that passes for marketable product. I know there are smart people involved in creating software, but they are not engineers and (with the possible exception of some groups at NASA[0]) nothing they produce is the result of an engineering process.
Has any software "engineer" ever lost his license because he signed off on trash code that failed? Where is the Otis elevator emergency brake[0] of the software world?
[0] https://www.eng.auburn.edu/~kchang/comp6710/readings/They%20...
[1] https://www.otis.com/en/us/tools-resources/high-rise-safety-...
> Ok so why is all their software such garbage, then?
Because the people involved weren't good at their job...? We've been over this multiple times now. It doesn't seem like it should be hard to figure out. What do you continue to miss?
> but they are not engineers
Sure, software involves lots of different types of work. Engineers are just one tiny piece of the puzzle. But design is exclusively the domain of engineers. Anyone working on the design is an engineer, by definition. That, again, does not imply that they know what they are doing, are good at their job, or bold enough to push back against out-to-lunch stakeholders. Having someone doing engineering work does not magically mean you are going to see an excellent product on the other side. Most people are pretty lousy at what they do.
You, again, increase your chances if you involve PEs, but that's not a job, just a designation.
> Anyone working on the design is an engineer, by definition.
You and I have very different definitions of "engineer". To me, an engineer is a licensed professional who adheres to the code of practice for his or her profession. We'll have to agree to disagree on the semantics. I think lots of software people want to call themselves "engineers" because it sounds like someone who should be paid more money.
...or it did, until the AI revolution.
> To me, an engineer is a licensed professional
Again, we’re talking about jobs (and whether they will remain in the future), not professional designations. Licensed professional engineers are usually referred to as PEs, but even if you want to call them engineers it is clear that not all engineers of that sort work in engineering. In context, we can establish that we are talking about engineers in the worker sense, not the professional license sense.
Is English a second language for you? Maybe context doesn’t translate well. English is certainly bad for assigning multiple meanings to words and relying on context to figure out which meaning applies.
> I think lots of software people want to call themselves "engineers" because it sounds like someone who should be paid more money.
How would describing the job by what it is, literally by dictionary definition, garner more money?
I am aware of the time physicians started calling themselves doctors to try and ride on the coattails of academics, so I think I understand the angle you are trying to come up with, but this isn’t like that. In this case engineer was already the word in the dictionary that existed to describe the work.
> To me, an engineer is a licensed professional
Even outside of "software engineering" there are tons of people working jobs that almost everyone would consider "engineering" that don't require a stamp. Chemical engineers, electrical engineers, mechanical engineers, tons. Only ~5% of chemical engineers have passed the PE, I guess there are essentially no real chemical engineers then?
Even in a civil engineering firm, most likely the kind of engineering you're talking about, only about half will actually be licensed and have a stamp. Tons of people will work on a project and figure out a lot of details, then work with the person with the stamp to sign off on the implementation details. Their work is undoubtedly still "engineering", but they don't have a stamp.
how many software engineers can reliably pick the failure modes of their software, and also build in appropriate safety factors on them?
Engineering is ultimately all about managing tradeoffs, so the more important question is: How often does reliably picking failure modes and applying appropriate safety factors actually matter in software, especially when most designs are little more than simple line-of-business CRUD with a (if you are lucky) slick UI?
>In contrast to most jobs, software engineering is unusually self-contained, with short feedback loops and immediate validation.
This is nice and makes things easier for the models to learn. Emphasis on easier. They are still capable of learning how to deal with other kinds of problems and being better at it, and they have, over the last couple years.
I'm a little surprised by how easy people joke about residential electricians and plumbers in this context. To begin with it requires young and healthy back and knees. How many men after 40 and especially 50 are like that and for how much longer?
Secondly, in CA to be a real residential electrician you need to be C-10 licensed. Which requires all kinds of time consuming prerequisites to begin with. Some of them could be a challenge if you are not the kind of guy who knows a guy. Apprenticeships, unions, depending on your major 1 to 4 year of "hands-on, full-time journey-level experience" and such. Your best hope is probably that your EE degree from 20+ years ago still counts for something.
And there quite a few 20 yo candidates to compete with who are taking DC/AC circuits classes in community colleges this very moment. They are not spoiled with life-long office work experience.
The competition among people who'd like to be an industrial robotics technician at the Tesla Fremont Factory is very real too.
This is generally how I feel - whenever new levels of abstraction have been introduced the goal posts for what can be built are now orders of magnitude further. This seems to be the case with all technology.
Consider assembly -> higher order languages -> libraries -> open source -> stack overflow -> now AI.
> If software engineering goes the way of the dodo
Much more likely is that software engineering goes the way of the elevator operator. Advancements in technology saw the number of elevator operators explode, to the point that we've all become one!
Unfortunately for those wishing to make a living as an elevator operator, due to the properties of supply and demand, us all joining the ranks of elevator operator means that supply is effectively unlimited and thus price cratered to effectively zero. This is the real risk to software engineers, at least those who are in it for the money.
The cure for high prices is high prices.
I guess the question is more if you're a coder or a builder. If software engineering was about writing syntax and flipping bits for you, this might no longer be an economically viable activity.
If it was about building useful or fun things, then AI will probably not take that away, even if it's just for the simple reason that you care more about it than some hot GPU.
Yup I agree. I do think that all the low(er) skilled software jobs will disappear. Those mostly working on marketing websites or simpler CRUD apps. And if they don't disappear, they will just not need nearly as many people for those anymore. But those jobs were mostly about writing code, i.e translating a Jira ticket to reality, and that's not really "engineering" as such, so if your only marketable skill was knowing how to turn English (Jira) into Syntax (Code), then yeah, better start upskilling fast.
I completely agree.
The engineering part of software was already completely dead for a large portion of coders who didn't want to know anything about the domain, left all the decision-making to business people/product managers/designers and was downright allergic to doing anything that was more complicated than a CRUD in their favorite framework (which would invariably either change daily or remain static for their entirely career).
There was a huge influx of people like that in the last 10 years who got into it for the money, and frankly the industry got much worse and much more hostile to different perspectives and different experiences.
People who have not evolved into builders have visibly become meat interfaces to the LLM.
Have any of these articles actually addressed HOW software developers are supposed to retrain for good jobs? Personally I have neither the money nor the years to go back to college to get a livable job.
I know some people whose work was already almost completely eliminated by AI (language translation) and they're not doing great. Reality is there's no-one out there looking out for you except yourself and very close family/friends.
I know a language translator (Japanese/English) who is doing very well out of AI screwing up and causing damage.
Oh yes, when LLMs fail at translating they don't make a visible error. They fill up with inferred meaning. Translators can only be replaced by LLMs if there's no consequence attached to the meaning of what's being translated.
This might be the savior for software developers
It could. But I'm not confident there will be a broad recognition of the fact if things become much worse than they are, or of the cause, or that there will be a desire to make things be better. There has been a long decline in quality in favor of optimizing for monetization and engagement, predating LLMs. But I do think there will be some pullback in things that actually matter if not in the SaaS FOTM.
kinda. there'd still be a lot less demand for software devs than there used to be, a lot of people would still be jobless/would have to pivot to some other career.
Just curious, what kind of jobs did they transition to? Or were they forced to take a break, or left in job application hell, etc?
Doing mostly language tutoring last I checked.
So obviously each college is different. And your experience is not mine.
But my college did not train me to get a job the first time. Indeed they explicitly explained to us that that was not their goal. That was helpful because up to then we did think they were doing job training.
What they actually set out to do was make us "smarter" people. In the sense that they taught us to analyze problems, work in teams, find creative solutions. They taught us about research and critical thinking, and logic and so on. How to hypothesize, test, evaluate, adapt.
You don't need to go back to college to learn all that again. You already know it. Skills are easily, and quickly taught and learned. That's the easy part. The hard part is approaching a job as a problem to solve- how and where to add value. Value people will pay for.
Look around you. See unmet needs. Figure out what you need to do to meet those needs. That's the real challenge we face going into this next era.
I don't know where you are, but when I hire mechanical engineers, I hire them from universities with accredited engineering programs. It's actually a legal requirement for the job in my jurisdiction. And if those engineers decide they want to become daycare teachers or knee surgeons, they do have to go back to university and re-train, even though they already learned how to think the first time.
You're right. There are several occupations which have legal entry requirements. You can't be a doctor without getting a medical degree.
However the number of professions that require a formal degree is very small. (And notably, programming is not one of them.) If you take any large group of random people and asked them, you'd find close to all are not required to have a formal certification.
Generally yes, people with (any) degree are more hirable to (any) job. The skills getting a degree are generally useful, and make a candidate more employable. But if you already have one, you already have that.
There are many fast-track teaching certification programs for people who already have a university degree https://tntpteachingfellows.org/
Healthcare is another growing field that's easy to get into with only 1- or 2-year certification programs for some positions.
Is that true for healthcare careers that don't have abysmal quality of life? All of the nurses I've known are working 60 hour weeks, night shifts, etc.
And get paid peanuts. And effectively they're told that they are disposable in the face of a pandemic.
Pharmacy techs work regular hours.
Good pay, low barrier to entry, pleasant. Pick two.
Not from the pharma workers I hear from in the US. But this is all anecdotal.
Did you graduate yesterday? Have you actually lived in the world?
My unmet needs are $2.5k rent and $2k childcare, per month. Then there's food, healthcare, car payments. There is pretty much zero wiggle room for us.
If you're unemployed, you won't need to pay 2 grand/month for someone else to watch your children, and you don't have to pay San Francisco rents if you're not working there.
I'm interested in how you expect someone to attend school/retrain without childcare. You can't generally bring your kids to class with you.
In some other countries family generally takes this role: grandparents look after them
I suppose you've got a point; if the middle class of Americans constantly have their careers destroyed, eventually we'll return to a third world sort of existence where childcare is provided by extended relatives.
However, most people in America can't rely on grandparents for childcare. One reason is we have children much older than people in the third world do - if two generations have children at age 38, it's not very reasonable to be expecting a 76 year old to provide child care.
Odd assumption that everyone has grandparents alive.
Fairly normal if having kids before 20 is normal.
And I suspect disaster to have correlated chances of killing parents or grandparents or children. i.e. if you've become a parent there's a good chance your parents did too.
Our own lived experiences are likely outliers, since developed countries have different outcomes than the past.
> if you've become a parent there's a good chance your parents did too
Crap and nonsense! Rewrite:
Once someone's survived to become a parent, there's a good chance the grandparents survived as well (assuming the norm is to breed ASAP).
Or assumes their grandparents are allowed within 200 feet of any children.
I don't think "just give up your existing community and move" is a viable option. That defeats the whole point since at that point the retraining isn't allowing you to maintain your quality of life.
It is a viable option. "Quality of life" is a luxury and programmers no longer belong to the class that can afford it. Do what everyone else does and go wherever the jobs are and live at whatever quality the market decides that you deserve.
I think you are missing the point of the discussion. The idea behind "just retrain" rhetoric is effectively "new technology won't destroy people's quality of life because they can just retrain!". If quality of life isn't maintained, then there is no reason to retrain at all, people can just get a minimum wage job and live in poverty.
The rhetoric was a lie. Capitalists lie. That new technology always creates a surplus of newer, better jobs and quality of life always improves for everyone was a lie. That you can "just retrain" and everything will remain the same was a lie.
Even if you "retrain" your quality of life will go down because the market value of your life will go down as automation and AI improve, unless you're a member of the capitalist class. Which, if you need to work for a living at all, you aren't. No matter what you do unless you're personally wealthy your quality of life will go down. That's just what an efficient market does to most people, and AI is going to make the market brutally efficient.
I wonder how many sectors this will have to happen to until workers realize that nobody's safe. I was shocked to see journalists at major firms peddling the 'just learn to code' line a few years back, given how much their industry had contracted in the decades before.
It's more nuanced than a lie.
Quality of life during the Industrial Revolution dropped considerably for most of the working population, but afterwards saw big improvements. You seem to be implying that life now is worse than before the IR. Are you? If not, in what way was the loss of jobs to mechanisation then more acceptable than it is now?
This is even more nuanced.
The overall quality of life improved through and thanks to IR. But the quality of life of the people most affected, and their children, and their children's children - did not do so well.
Technological progress automating away jobs lifts humanity as a whole, but it absolutely fucks over the people whose jobs got automated away, and the families of those people, and the communities they live in.
San Francisco is where the jobs are, and dropping your entire social network on the floor at exactly the moment you need it the most to help you is not good advice.
Former tech workers can ask for advice from non-tech people who previously had to leave because they could no longer afford rising costs that happened due to influx of well paid IT jobs.
San Francisco was where the jobs you want were but now those jobs belong to a Claude account.
True.
> just give up your existing community and move" is a viable option.
That’s been how humanity has always worked, human migration is probably the largest single driver of social change in human history
Millions of people migrate every year today, I’m not sure what you think the alternative is
No, it isn't. This is an incredibly modern phenomenon. We see examples of peoples moving en masse throughout history, yes, but the idea that everyday individuals should expect to leave their community is an incredibly recent one. Such would have seemed monstrous to people even a few hundred years ago. Fun fact: Socrates wasn't actually sentenced to death, he was sentenced to death or exile, and chose death over having to leave his home behind.
> Millions of people migrate every year today
A just society that doesn't force them to.
The entire colonization of the planet by humans from 250,000BC to today would argue otherwise
We’re physically built to be mobile - the extent you can take your community with you has always been relevant
I’m sitting in Virginia USA with 50% of my neighbors being migrant families of 5+ people usually across three generations
I’m not sure what data you’re working with. In fact there a whole lot of ignorant politicians that want to stop migration precisely because of how valuable of a tool it is.
Some humans moved their families from somewhere that they weren't safe, to somewhere that they could be safe, with the intention of settling there and building a living space their great-grandkids could be proud of.
Some humans lived nomadically, moving all the time, as a way of life.
Being forced to do the second because nowhere is stable enough to do the first is what we're talking about.
> The entire colonization of the planet by humans from 250,000BC to today would argue otherwise
That colonization happened when there weren't other humans there. Later colonizations were either unsuccessful or led to very nasty events.
Also, today jobs are concentrated in places where things are expensive.
>The entire colonization of the planet by humans from 250,000BC to today would argue otherwise
You're talking about time scales on the order of several thousand human generations.
Take any human over the course of the last quarter million years, the chances are they didn't migrate away from their communities.
Migrating is one thing, having to migrate another.
-- from "Home" by Warsan Shire, https://www.amnesty.ie/wp-content/uploads/2016/06/home-by-wa...I'm assuming you're talking about people born in SF? Because the rest have already demonstrated a willingness to give up their community once in the pursuit of quality of life.
Equally, if you are in SF you're about to feel the pain you've caused to the locals by moving there. Somewhat harshly, I could suggest that you didn't care about their existing community when you priced them out of their neighborhoods.
So here's the rub. SF is somewhat different. It is a company town. If you don't work for the company you're moving. Sorry. There's certainly nothing else that can pay you the same money.
On the up side, there are lots send lots of cheaper places to live. And chances are you can find a good job there doing something that adds value to that community.
Your solution to childcare is permanent unemployment?
This is Denver rent actually, and increasingly most urban centers.
That's a big part of the problem in these discussions.
Those graduating today have plenty of buffer to reskill tomorrow. Those of us who graduated 20 years ago, have no chance.
Only $2.5K rent? That's like free these days. For what though. I rented a 4K sq ft house in Charlotte three years ago and now paying $3.5K for not a modern 2bdr in Miami proper.
1100 a month in Kansas City and $800 a month in Denver. Everybody wants to live on the coast.
$800 in Denver for what, a studio? A closet?
1br in Denver was well over $1000 last i looked.
Yeah lol $800 is like a bottom tier 1-room apartment or nearly derelict house where I am and that's a blown out rust belt shithole city.
The Gulf Coast?
> I rented a 4K sq ft house
Excuse me, but what could you possibly do with a ~360sqm house than one half the size wouldn't do? A family of 4 for sure fits in even a smaller one.
Because that's how they are built in the US. I didn't need the extra space, but for the money, there wasn't much choice, really. I would have loved to rent a house half smaller at half price, but none were on the market. I barely went upstairs. My dog occupied it, at least.
I could have rented a townhouse for $300-400 less, but then there are different constraints. Small or no yard, tiny kitchens, no garage etc.
I was perfectly content with a 2200 sq ft ranch in Henderson Nevada.
> Because that's how they are built in the US.
That's just sad. I knew that real estate was a pain in most countries around the world, but I never realized that choice in the land of freedom was that constrained. I guess in most places it's basically: "Detached House, Large".
The Charlotte house was (2022ish) $2.3K/month through a PE that bought out a lot of single-home houses to rent out. 2 story, 4bdr, 3 baths, a yard and a lawn, a huge deck. The area was very nice and so were the neighbors. My dog loved the deer.
It wasn't urban enough for me, for someone working remote at the time, and few growth opportunities for my wife, but I regret moving to FL. Miami has become the least affordable city in the US relative to income.
For sure! Still more money than I can fathom bringing in if software engineering went away tomorrow.
1300 sq ft house with a big yard for the toddler.
I keep telling my 22y who graduated from a university last year to go to a trade school (HVAC/ electric, plumbing). Too late for me as these require apprenticeships in the US.
My HVAC contact in FL needs a $100 just to sniff the AC/boiler closet. That is, if I can get him to show up. I don't even know if he's licensed here, but he's definitely very experienced.
This works if you're interested in running a business, but what well paying work doesn't require a degree or some kind of formal training certification?
There's a lot to unpack in your question.
First you assume that being "interested" in the work is a requirement. I'm guessing 90% of people are doing work that doesn't interest them. But assuming that it's a requirement for you the question becomes, what are you interested in?
Second you want well-paying work. I understand. Programmers got well paid. They lived above the standard of most. It's not surprising you'd want to keep that. But "well paying" is a spectrum. It depends on your life-style, location, eating habits and so on.
And finally, lots and lots of jobs don't require a formal certification. Or one that isn't trivial to get. Ask the people you see everyday (including the software programmers, hardware designers, receptionists, assistants, book keepers, marketing and advertising people, sales, support, your vehicle mechanic and so on.) I suspect if you asked every dingle person you met in a day the ones that actually had this requirement would be a rounding error.
People don't get a degree in "generic doing things people want". They generally get one focused on a particular field.
If someone's degree is in computer science, they're not going to be a good fit to be, say, an accountant or a nuclear engineer.
I know what you’re trying to say, but strictly speaking there are many examples of “hard-science” people moving towards and excelling in softer fields, while the opposite is almost never true.
There was a saying where I went to high school: “the language programme students are not better at languages, they’re just worse at math”.
The language students were however better at writing and communications in general. They ended up in sales, marketing, advertising and everywhere else.(all under threat from AI as well.)
But to further the GP question there are lots of examples of people not working in their degree in both directions. The entire computer industry was built by people without a Computer Science degree. Today huge numbers of programmers have no degree at all.
To refer back to the original post about the farrier turned mechanic. Most comments seem to miss the point. Change is coming. You can lean into the change and prosper. Or fight the change and vanish. Those with enough vision, creativity, adaptability will continue to thrive.
The same way clerical and manufacturing workers retrained when their jobs were replaced by computers/the Internet and robots. That is to say, most won’t.
Well, a lot of them retrained into call-centre type of jobs.
Then the call centre jobs offshored (and now are being replaced by AI).
People in tech change roles and careers all the time. The vast majority of role or career changes do not involve going back to college.
I would even recommend against going back to college unless someone has an extremely compelling case for a career change that requires it and the conviction to see it through.
I'm hesitant to give examples because I know every example I give will be met with objections from "But I don't want to do that job" or "What if that gets replaced by AI too?". Look around you and you can should be able to identify some adjacent roles where you at least have an understanding of what the works like and you could learn the rest by taking on gradually more responsibility.
There are a lot of jobs out there that aren't exciting, don't pay SWE level salaries, but nevertheless pay enough to support a family as evidenced by all of the people doing those jobs who support their families. Not every company or industry is interested in squeezing maximum efficiency out of their employees. I have friends in highly bureaucratic roles who care about their jobs and do it well, but their work is amazing simple and easy. It requires nothing more than being a responsible adult who can show up to the office and get things done. These jobs are everywhere and I don't think they're going to disappear.
I don't think SWE is going to disappear either, but it will look a lot different in years to come. If you can't or won't adapt to the new normal, one of those boring alternative careers might be worth pursuing.
>The vast majority of role or career changes do not involve going back to college.
That was possible pre-2022 during the famous developer labor shortage. In the current market, companies expect seniority/experience in that specific stack they use for that, no more career changers.
They either take you as a senior/expert to contribute from day one, or as a young and cheap newgrad/junior out of uni they expect to train on the job, but no more in-betweeners switching careers as they're not as experienced as the seniors on the market, nor are they as cheap and malleable as newgrads/juniors.
Sure, in theory every developer can learn at home to do anything new related, but in the current job market, companies expect X-years of proven on the job experience with their stack, no more bootcampers, and no more self-taught-at-home types. The entry bar has risen significantly. At least where I live.
Like I just saw on linkeidn a job posting from a large bank that was an hour old and it already had 100 applications, which probably means most applications are now AI botted, meaning yours will get lost in the noise and getting a call back for an actual interview is like rolling a dice.
Not saying it's impossible to change(I did a couple of times), just that's it's much more difficult than in the good times, and will most likely also come with a paycut.
To make matters even worse, if you do apply to these jobs the only thing you get back in return is your data being sold to even scummier organizations. I always use single-purpose emails for everything, so I know where things source from. Imagine my shock when a week or two after applying for jobs I started getting a steady stream of Super Intelligence emails offering to spruce up my resume, or to source leads for me, or some other sketchy service.
Everything is commercialized now. Even the discard pile of resumes can be sold for pennies. A job application isn't a job application anymore: it's a desperation signal alerting predators to come extract some value.
I'm glad to be protected by GDPR.
How? I'm still getting scam calls despite GDPR, because the scammers in India don't care about GDPR.
I guess where I'm coming from is that I currently am trying to change careers. Everything I'm finding is either a 50% pay cut or is simply unwilling to consider anyone without 5-10 years of experience in that particular industry. Since I've already built a life here I can't really just move and a 50% pay cut isn't enough to live anywhere near a comfortable life; particularly since I have medical expenses that aren't covered by insurance.
> I have friends in highly bureaucratic roles who care about their jobs and do it well, but their work is amazing simple and easy. It requires nothing more than being a responsible adult who can show up to the office and get things done. These jobs are everywhere and I don't think they're going to disappear.
I think these jobs are actually the most likely to be replaced by AI. They have resisted automation to date because there's too much fuzz in the inputs/outputs to make it feasible to computerize. LLMs solve exactly that problem.
Except for the part where they have absolutely no idea what the world is or how it works. They are good at software and math because of RLVR, but they can’t run a business.
>They are good at software and math because of RLVR
Software and Math are not the only things LLMs are fast improving on. Astra is one of the biggest examples of this being untrue. The model is cracked at computer use and vision/spatial tasks and reasoning. Driving, Operating Robots, Video Editing, 3D modelling, graphics are all things Astra was >>> at than any other model.
>but they can’t run a business.
And it's the leader on VendingBench. Also most people cannot run a business.
Kinda true (they have a mediocre idea of how the world works), but it's surprising how close they are given their general limitations and what it took to get disruptively good at software and maths.
Like, sure, using an LLM as a CEO today will bankrupt you and some of the decisions leading to that are daft, but even despite that it's in the same "huh, interesting" kind of zone that early LLMs like InstructGPT and the first release of ChatGPT were with code. It could write short functions or small scripts for you, even then… but any attempt to go beyond that would fall over quite fast.
Astra told me I can walk to the car wash to wash my car.
Again, the improvement in coding is due to RLVR, and it does not generalize to other domains. So it is meaningless to point to examples of that improvement.
There are many ordinary behaviors and skills that are not feasible to RL at scale. The sampling efficiency is just way to low.
> Astra told me I can walk to the car wash to wash my car.
Yes, like I said, a mediocre idea of how the world works.
> Again, the improvement in coding is due to RLVR, and it does not generalize to other domains. So it is meaningless to point to examples of that improvement.
I can point at actual benchmarks for performance on business tasks. On the one hand, those lines go up; on the other, I just don't know how good those benchmarks are.
> There are many ordinary behaviors and skills that are not feasible to RL at scale. The sampling efficiency is just way to low.
Yes. A point I make myself on occasion. One of two perfectly valid ways to describe intelligence, and one by which all current machine learning really sucks.
Like, if it was fully general, the trillion or so miles that Tesla cars have driven would have obsoleted the steering wheel long ago, rather than the closest we have being (yet another) pronouncement from Musk that this new car he's not actually made available for sale won't need one.
Business self-play training could be anywhere from code to cars. I was expecting much faster progress with the cars, and much slower progress with code, so I don't know how to adjust my expectations with any other domain, including business.
It’s much harder than both code and cars because there is both more variety to the work and much work where the reward function takes months or years to resolve objectively.
In general it could well be as you say, but the current failure modes are rather shorter-term than that.
I won't be surprised if current approaches can get decently good at e.g. managing a corner store or a restaurant, without scaling up to a 1000 person R&D firm.
(Famous last words).
The trick is not to start in software and have something to fall back in to.
In my case that’s mathematics in academia so I’m going to be starving to death.
I started from sweeping floors and making survey calls at minimum wage. I think you can teach Math to high school students.
Have you checked how much high school teachers actually make? It's not much.
Still better than sweeping floors definitely. And you can work as tutors if you are good with kids.
Actually that's not generally true. Janitors at least work 12 months a year, many schoolteachers are paid ~9 despite being expected to work through them (on lesson prep &c).
you know I would go back to cleaning toilets if it paid the same as staring at a screen
It's possible to make low 6 figures, but a lot of pay is dependent upon length of time teaching and/or which district to get into. Good districts rarely have openings.
I'd say (not a teacher myself, but related to several) teaching is a lot more difficult than people give it credit for.
Just like everyone else. I have a feeling your definition of “livable” is not in line with the reality that most people face. You share more in common with a stay-at-home mom going back to school to become a phlebotomist than you think. It’s just pride gets in the way
A stay-at-home mom isn’t bringing any income in so has the luxury of being able to fail.
Whereas many of us can’t take a loss of income for months while we retrain.
You don't have any buffer at all? You're spent your entire inflated software developer salary for the entire time you've had it?
Not enough buffer to take 3-5 years to rebuild a career is not the same as no buffer at all. Personally I have about a year of buffer which is no where near enough to meaningfully retrain.
Why are you assuming every white collar worker has got an inflated salary, or that they’ve had it for a while?
And actually no, I can’t afford to be unemployed for a year. On COVIDs cost of living it would have been easy. But the current cost of living I’m not able to save even with my “inflated” salary. And the savings I do have isn’t going to stretch as far.
I honestly don’t know how people save on lower incomes. I suspect they don’t. Which is why the debt crisis keeps worsening.
I’m lucky enough to have assets but once they’re sold they’re gone. And most people don’t even have that luxury.
I think you're ignoring risk.
Even if you've saved enough for 1 to 3 years of university, there's a risk that you won't get a job especially if you're getting towards 40+.
Drawing down savings is a very high risk move, since you could easily never recover those savings because you may struggle to get a job that has a better ability to save as much.
Maybe they're spending it so their kids can stay in school instead of going to work.
I've been in this industry barely five years, 401k tax penalties are huge (and that's where a lot of the savings is), and rent is expensive
[dead]
I don't get it. Didn't you live way below your means so you can save up money for retirement? I've got enough money saved to get away with being jobless for about a decade.
Not everyone's had the luxury of saving for years and years. Some people just got into the industry (remember 'learn to code'? not to mention recent graduates), others had to work their way up through lower-wage roles (e.g. contracting) just to get to a place from where they could hope to save. Other people have family to take care of and can't just throw everything into their savings.
Even back when I had to make do with $400 a month I was saving. Live out of a van to save on rent if you have to. There's always a way to live below your means and save.
Of course there are exceptions, but I don't expect those to apply to anyone commenting here.
A bizarre perspective - you seem hung up on the virtue of saving but totally disconnected from the material implications - even if you saved 50% of your income for a year at that time, you’ve finished with $2400 in the bank - how long is that going to float you? About a week in my case.
It’s plainly true that for a significant number of people, even those who satisfy your moral sensibilities around money, they won’t have had sufficient time at sufficient incomes to have enough saved for a significant period of unemployment to retrain.
If you burn through $2400 in a week, there are lots of opportunities to make changes in your life to make do with less. I'm currently living off $240 a week and that includes going to a fancy restaurant, so I could cut that number down even more if I had to.
You have no idea what my circumstances or expenses are, and in fact they’re irrelevant. You seem still to have a strange inability to see beyond your own circumstances.
You told me your expenses are $2400 a week. I told you that it is easy to get this number down since it's unlikely that you have to pay $2400 a week in life extending medications.
I'm not the GP but I will answer the following on their behalf:
It's usually best not to make assumptions about other people's personal circumstances.
Living out of a van is illegal here and good luck getting a job with no permanent address
Live out of a van with children? Is that what you're suggesting people do? That's monstrous. You would rather put the weight of the world on a poor family's head than one iota of responsibility on the ruling class.
Since that time living out of a van has become a crime in a lot more places.
And do you feel comfortable drawing all that down so that you're left with nothing at the end of the decade?
Is that in cash, so if the stock market drops 80%, it's still true?
If the stock market drops 80% and stays there for longer than a year it's Great Depression 2.0. Retraining won't do jack shit. There'll be breadlines, governments paying people to dig ditches and build trails in national parks.
Lately I've been just thinking of hitting ALT+F4 on life for good, as a big FU to the megacorps (not that they would care, but it would give me the satisfaction at least)
There’s so much more you can do to get back at megacorps that will hurt them more than that. 1 less person in the world to oppose them is a win for them.
I’ve never met you but based on your last 10 HN comments I think the world would be worse off without you in it.
"as a big FU to the megacorps"
Where is the satisfaction in that?
If you jump from a bridge, all it does, is that relatives of yours will receive the bill for cleaning up the blood and remains (on top of funeral etc).
Just ALT+F4 on shitty tech, which is most of it. Ditching your smartphone is a small but good start.
If you're going to do this at least take down a data center or something. Most new constructions aren't insured at all.
I think of doing this except for suicide (which you shouldn't do) I fantasize about becoming a monk. That's a life role that can't be replaced with AI, provides meaning to yourself and to the community and guarantees you a certain amount of happiness – probably more than most jobs.
Monasteries are wary about people inquiring because they can't hack it in the world. And because monasteries are businesses of a sort, they are concerned with what a middle-aged inquirer can bring to the table financially.
That's a very broad statement; the reality depends on the specific order and monastery in question. To be sure monasteries are wary about all who inquire, thus the usual two-year neophyte period. Beyond that some will expect more from novices, other less, but there are many stories of broken people retreating to monastic life even today. Indeed many (within the Church) expound that as a key role of monasteries, providing a place for people to escape from the world. If you just want to live a Godly life, you can do that in society; monasteries are precisely the place you go to do so outside of it.
A pithy saying in both the Catholic and Orthodox traditions is that people shouldn't consider monasticism if they want to run away from the world, but rather only if they want to run towards Christ. A significant part of the discernment process is weeding out the former.
[dead]
> hitting ALT+F4 on life for good
That hardly seems like it would be satisfying.
Getting permanently free from all this nonsense that plagues human societies wouldn't be satisfying to you?
It wouldn't be satisfying to you either, as you'd be dead.
same, not like, for megacorps, but, in general, too tired of this plane of existence, and, nothing really interesting/exciting?
There are many ways to do that more effectively TBH.
Only if you go out with a bang
You retrain the same way that we told workers when we decimated the garment and manufacturing industries. You'll get the same amount of government support too.
Who's we? Personally, I've actively supported the traditional garment industry to the degree that it still exists.
It used to be you could just buy (for example) jeans made in America for $20. I bought such a pair back in the late 1990s. There was no need to be busy trying to "support the traditional garment industry".
Nowadays you'll be spending $40 (same price after adjusting for inflation) for something definitely not made in America, and probably of inferior quality too. Since the domestic garment industry has shrunk so much, the supply chains an economies of scale don't exist anymore to buy domestically-made garments of decent quality at those prices.
As recently as the 1980s-90s (when I was a kid) 50-70% of American clothing was made in the U.S. Post-NAFTA, that declined to 3-10%.
We were also told it was important to help create good jobs in Mexico so there would be less desperation resulting in mass migration from Mexico to America. Good plan, right?
Well, it was, until the factories in Mexico also closed as the jobs migrated to China, Vietnam, and so forth. So Mexico got to enjoy the same problems America had, except they were poorer to begin with.
That's my point. You are alone and your government doesn't care.
No - because as a software developer adapting IS the job.
Exactly. Though this round of tech change is much faster than the previous ones.
I would like to know the answer to this, too.
Not really sure what I should be doing on the side since AI seems to be automating a lot of work that I do as an analytics engineer. Maybe in 10 years all I'm doing is feeding in markdowns and project specs to an LLM and reading test outputs. At what point are you just a technician?
The bright side is that analytics engineer is a relatively modern job title, previously it was largely bundled in with data engineer, and before that, data scientist. So you're either already relatively early in your professional career or have adapted between programming disciplines once already.
I also (personally! I'm in data engineering for my part) feel it's one of the safer roles from an AI perspective. Yes, you'll be writing a lot less code in the future, but that's never been the particularly difficult part of your job: SQL is dead easy to use.
The difficult part is picking up all the business context and knowledge in order to work out the best ways of representing complicated and messy data for the business. AI can help to some degrees there, but if you're great at your job, your judgement is going to be built on so much business context that both can't be written down and would be too vast to feed into the model.
Sure, AI can do some of the job, but you'll always thrash it in terms of quality (whilst getting to spend less time doing the busywork of just writing your thousandth predicate of the day!). And worst comes to worst, all the business knowledge you have in your head makes moving into analytics a far more viable lateral move.
And that one is definitely AI proof. If all companies use AI for their analytics, they're going to be moving largely in lockstep with each other. Humans might be far worse, but often far better, so I'd be willing to bet that through sheer randomness that over time companies using human analytics will outperform those relying on AI.
We are mostly already doing that now as DE.
I have a friend who is a skilled UX designer and UI artist.
He works at Best Buy now.
Wonder what his next job move is gonna be able to be.
[dead]
well, I suppose the same way that other people whose jobs were lost due to automation over the years were supposed to retrain for good jobs? I doubt they had the money or years to go back to college to get a livable job either.
In the past there was always demand for "unskilled" (manual) labor where displaced workers could land to without long (re)training, and possibly even gain skills on the job. But that job marked has dramatically diminished especially in the west; there are only so many burgers to be flipped. Somewhat famously during industrial revolution displaced agricultural workers (e.g. farmhands) ended up mass-migrating to cities doing factory work under extremely exploitative conditions. And that never truly got resolved, we just brushed the issue under the rug by pushing manual labor to third world countries where the problems could remain out of sight.
Generally speaking they didn't. They either got lucky or their quality of life plummeted.
hence why I put "supposed"
Hence why we had things like the rust belt with excessively high drug use.
You're just not creative or entrepreneurial enough to see the opportunity in it. Just become the drug dealer.
"Vibe-coding rescue" is a new industry. Check it out.
In my experience, AI has given far more leverage to software engineers than anyone else, because they have a feel for what computers can do, what's easy, what's hard, what's performant, or poorly specified, etc. Therefore I think retraining is mostly about being willing to experiment with AI-assisted workflows and not being too tied to old ways of doing things. As long as you are focused on the problems you are solving, I think you'll do okay.
What I'm more concerned about are the younger generation where all the learning tasks have been automated. I have faith that the strong, naturally curious engineering types will still learn to do great engineering, and honestly without some of the mental blocks and calcified assumptions that us greybeards have. But how do they get in the foot in the door and the reps needed to develop senior-level judgment is another question.
In some sense we need the current mania to die down and the long-term maintenance implications of heavy AI assistance to become more apparent. Basically I think we need the AI bubble to pop and then 1-2 years for things to stabilize, and then we'll recalibrate on hiring expectations and best practices for AI usage. The current AI maximalist and doomer views I think are both basically driven by capitalist incentives and wishful thinking from investors and a reality check is going to come in some form (whether technically, socially or politically).
You are presuming that you will need to retrain for a new job. Why do you think that?
Because I've been looking for a new job for several months and all of them either require significant retraining or don't pay nearly enough to maintain even a similar quality of life. I don't just mean they don't pay enough for luxury. I mean they don't pay enough to live in my community without a roommate.
Also what job do we retrain for?
Many of us on this site have worked directly on projects where the efficiency gains ultimately reduced or eliminated positions. If you want to exit the development field, I think the answer is probably the same as for the people whose jobs we helped eliminate.
The internet has made it easier than ever to learn new skills. You don't need college unless you're targeting a career that requires, or expects, a degree.
"unless you're targeting a career that requires, or expects, a degree."
Unfortunately most well paying jobs do.
Maybe there will be an inversion. Lots of white collar office jobs can be partially automated, or have their efficiency greatly increased by AI. But we still don't seem to be near widely deployed robots that finish drywall
There may be an inversion as white collar jobs sink, but not because drywalling is going to get better. All of my family work in trades and it's a lot harder and less stable than we office workers like fantasize, as much as I love the ending of Office Space too. It's cyclical and sometimes there's not enough work. Cost of materials is volatile and can eat into your income. And it's skilled. Programmers tend to think everything else is easy and if you can teach yourself React then electrical should be a no-brainer. That's to say nothing of start-up costs. It was incredible that we coders could bootstrap six-figure careers with a $4000 MacBook Pro. Trades workers need a truck, you need power tools, if you're not loosey-goosey then you need insurance. All of that costs a lot.
That's for anyone joining the trade. Out-of-work middle-aged programmers with soft keyboard hands and little or no experience hanging dry wall? Why pay them any more than the 24 year old guy. AI may automate us out of our desk jobs, but the downward pressure of us transitioning to trade jobs, even at the bottom level, will only make those jobs less lucrative.
You think the drywall company is going to hire you? Why would they hire someone in middle age with no experience when there are no shortage of 20 year olds who already know the trade?
Also, illegal immigrants, who are actually the people who do most of the drywall.
Do you like working with dry walls? I don't. I can do it, but I ain't born for handywork.
Interviewers and hiring managers make that decision. I'd say the better paying jobs require a degree or training.
[dead]
Pretty much none of us are doing jobs that existed in colonial America.
How can you not see that this time its actually different? The very need for labor is being obviated. Every technological change we had before still required a need for labor often an massive need that justified an entire city population. That is the difference. For the first time in human history, economic output is being severed from requiring maintenance of a labor pool. Considering we are that labor pool, things do not bode so well for us. As you state, look at the relative lack of abundance of candlestick makers and horeshoe makers and basket weavers and other colonial jobs. We are essentially making ourselves extinct and we don't even realize this.
> this time its actually different
The older I get the more time I heard this. Never actually true. But always used by somebody asking for power, to protect me - of course.
There has never been a technology with the promise of obviating both knowledge and physical based labor before. Now there is. Sorry, but this is unprecedented. The hubris to imagine all will just be swell is astounding to me.
They said the same thing when the looms put weavers out of a job.
The loom still needs people to run it, you fool. This one doesn't. How are you so blind to this nuance?
Then everything would be free.
There's an off-color joke that could be made here but the real question is the speed of change - we gloss over it but there have been major changes that took centuries, and others that were quite fast, this one seems to be closer to the fast end.
Yes and? You'll be shocked to know that pretty much none of us were alive in colonial America.
We are all going to learn the hard way, this is what the economy is like for many people. Always on the precipice of losing everything and thus willing to do anything like the capital owners want it.
Whatever cushy career path you think is safe now will probably not be so safe by the time you go to university, train, land a job in it etc. The job of the future is nurse, diaper-changer, cleaner, prostitute for elderly retirees. Welcome to the aging fascist AI technoutopia.
[dead]
Okay, I'll write one for you...
https://soj.ooo/p/positivevibesonly/post/f467feb1453208a492d...
And my great great grandpa thought horses would lose all their jobs to cars. And they all did and it was never the same again.
There’s no quick answers, but intuitively I think we all get that the % of jobs not doable by AI and robotics approaches 0 over time.
As someone that works in robotics/automation, it pains me to see people in creative/office verticals getting all the shit with AI these days while hard labor positions, the one that everyone could be doing a bit less because they take a toll on you (working the fields, farms, hazardous tasks) remain pretty much the same.
don't we already have lots of machines that do things like this though? Even if its human operated: excavators, combines, tractors, and other heavy machinery have reduced the physical burden involved in sustaining large societies tremendously.
Yes there are fewer hands on the job site but there are still hands at all. This is increasingly becoming not the case in the future. The goal is to fully eliminate all hands and all minds. There will be no more sponges of newly available labor. Then what happens exactly no one really knows or even wants to meaningfully try to answer, because the implications are too disturbing.
I would agree with you, if we taxed the ever-living fuck out of AI companies to enable UBI, so that the people replaced with robots did not end up homeless.
Well, first these AI companies would have to turn a profit.
Ideologically I don't consider myself to be a communist or even leftist, so I'm confused by my own reaction as well, but maybe mandatory joint-stock where the government takes a large chunk of equity of any AI enterprise off the bat and manages it like a sovereign wealth fund is the right answer.
No thanks. The last thing I want is the government with even more motivation to shove AI into everything in order to make it "profitable".
If you use a computer to do something, training an AI to use the computer to do the same thing just takes the computer, which was there anyway. (Yeah, plus training the model, that's not my point.) But if you do something physical, replacing you takes something physical to be there.
We may eventually manufacture the robots to do all these things. We aren't there yet.
How long do you give it?
I looked into this (or rather, had Google Deep Research do a 30-page report on it!) and it's still a very long way off. So long, that the AI, or any reference it found, would not even give me a rough timeline, but I suspect it is multiple decades away. The main blocker is a very severe crunch in the supply chain of critical materials, mostly rare earth elements which are essential to make those high power density + high precision motors feasibly.
Today we have a tiny fraction (like 2%?) of the global supply needed to make enough robots to replace all humans.
This is before we get to the other logistics like power and the rest of the materials, but we have a path to scale those up (except maybe Copper, that's another tough one.) Turns out rare earth elements are actually very abundant, but extremely hard to mine, very toxic to extract (which is partially why China has a strangehold on those) and we don't have a good path to scale those up.
All of this could be moot if we figure out alternative technologies for equally capable motors that could be manufactured at scale, but AFAIK whatever promising is on the horizon would still take a very long time to ramp up.
That doesn't make intuitive sense to me.
First, you make a robot that mines rare earth metals. Then construction robots. Then an automated factory for assembling robots. The whole premise is that the robots are faster at these tasks than humans would be.
If the robotics are feasible their rise should be exponential, not linear, since each robot built makes building the next faster and helps to remove whatever bottleneck you had.
Mining is not limited by human labor, we already use huge machines (and some robots) controlled by a few humans. The limitations are investment in huge machines and processing plants, and regulations on where/how you can mine/process.
You can't just have a robot start mining/processing rare earths, anywhere in the country. You need permits, and a lot of money. I guess that didn't stop AI companies ... but there isn't an "infinite" novelty/hype premium for rare-earths, like there is for LLMs.
> If the robotics are feasible their rise should be exponential, not linear
That totally does not follow. Robots can be feasible and their payback curve can still be linear or sublinear.
Take mining rare earth metals, for instance. The problem with mining rare earths isn't that it takes so much human labor. The problem is doing it without poisoning the surrounding 30 square miles. Replacing the driver of the earth mover with a robot doesn't fix the chemistry at all, nor (I suspect) does it significantly affect the economics.
We're waiting for the point that replacing the humans is the cheaper option. Not sure that robot plumbers are in our near future.
What's your definition of "near future"? 3 months? 1 year? 5? 10? 50? 100?
Not the other commenter, but we won’t have robot plumbers in 10 years. Maybe in 50.
Robots have been used in industry for decades already. Most jobs won't be worth investing in the robot doing it, since human labour is cheaper than the capital investment.
To the start of it? Maybe two years?
To the end of it? Maybe 40?
Worse, it was just almost all horses. There are a few in Central Park pulling tourists and that’s enough for the Pollyanna’s to point to and go “See! We’ll be fine!”
Horses where the plaything of the rich. Farmers were using oxen not horses for the most part. However the modern plow (invented in the mid 1830s - John Deere being the best known inventor, but others were inventing them about the same time) needed the faster speed of a horse to work well, but if the previous farming technology worked for you then oxen (or donkeys ... - there are several other work animals used in different areas for farming) were better.
There are likely more horses today than there were in 1900 - but now they are play things for the upper middle class living in the exurbs (commuting via car to their city job). Sure there are a few Amish (and other similar groups) making a living with a horse, but most of them are toys for the wealthy.
Do you have a citation for the horse number? My research [0] shows we have a 1/5 out peak number in ~1900 and our population has 5x since then. This means per capita horses in the country is down 96%, while absolute count is down 80%. Both are huge numbers.
0 - https://datapaddock.com/usda-horse-total-1850-2012/
The page you linked doesn't count recreational horses.
Good call! They have info on general population which shows about double the number of horses I originally thought (at least in 2016), and declining at a good clip of 10%+/decade) . So drop is still large but just a bit smaller.
The general point is before cars horses were used a lot, entire industries were based around them and our now gone. The toy factor does not compensate because of cost and inconvenience of owning a horse.
0 - https://datapaddock.com/
The number of horses in the US is down by about a factor of 3 from the peak in 1920. They're mostly recreational now though.
Presumably you are a human, not a horse, yes? Presumably there is more to your life than dragging around carts all day every day?
HN: A site full of bold futurists where it's easier to imagine the end of the humanity than the end of capitalism.
> the % of jobs not doable by AI and robotics approaches 0 over time.
Horse are an analogy. When they were more economically useful, there were about 5x the number of them. Now they are basically pets and we get comments like this suggesting the work they do is virtually meaningless.
Yes, most of the work many people do is virtually meaningless.
Are you a preschool teacher? Are you a doctor, a therapist, an artist? Does your work contribute anything to yourself or other humans around you?
Or do you get paid a lot of money to sit in an office and update JIRA tickets about tracking ad conversions and increasing engagement OKRs?
> Are you a preschool teacher? Are you a doctor, a therapist, an artist?
We've already automated the artists. Doctors are currently a strong area of focus. In a world where AI replaces human thought, education becomes only for hobbyists. That leaves therapist.
I suppose we're going to need a lot of therapists.
LLMs have much better cognitive empathy than most humans have limbic+cognitive empathy - the therapists will also be automated - friction like licensing bodies will slow it down but not stop it.
I’m not sure how to respond to this before I know if it’s satire or not?
Unfortunately not satire - modern models can be trained to be better than humans in recognizing emotions straight off camera.
That is not even 1% of what therapists do. Let alone medical ones/psychiatrists.
People like that always have such dismal opinions about what people actually do, it's insane
recognizing emotion is not the main thing for therapy, having a human on the other side is the only essential
You need one if you think engineers have automated the artists.
This may shock you but there are still in fact humans who drag around carts every day
Because the capitalists have very big guns and a monopoly on violence.
Its very easy to imagine a better world.
Its much harder to imagine the Scrooges in power willfully giving it up so that we can build that better world.
It's not just the Scrooges in power. Previous attempts to actually make a better world often made a worse one, and not just because of capitalist opposition.
If you mean attempts like centralized communism - absolutely - and the failure mode should have been obvious from human nature: centralizing great power incentivizes corruption.
We don't fully know the failure modes of every alternative yet, but that doesn't mean powerful central governments + capitalism is the best alternative.
Interestingly, the Amish shape their lives around this sort of thing. With any technology, they ask themselves, "Does this make us dependent?" If the answer is yes, they refuse it.
They are more and more adopting modern technology. And where they haven't directly they have found ways to use it and be just as dependent while appearing to not. Go visit one and you discover they want to borrow your phone to buy things from Amazon...
Uncle Ted warned us about this.
It's okay, I'll just use AI to vibengineer a viral basilisk that only targets those who violate the NAP.
This is the Chinese century, and even if China charts a somewhat different course than Western capitalism, it still believes that human beings should spend most of their days in some employment, even if it is just makework. A population with mostly leisure time is viewed as very threatening to social stability. I can see that model being exported to the rest of the world far more than I can imagine your utopia.
Looking at the age structure of the Chinese population I have doubts that it will be "their century".
Have you got a citation that the Chinese government believes in makework?
The Chinese social safety net, for individuals not covered by unemployment insurance, is strongly centered around public sector jobs (公益性岗位) that, if you see them in practice, are often plainly makework: the essence of the work has been mechanized/automated, but human beings will still be directed to be somewhere all day and look busy. There is no paying those people a dole while they stay home like in sone other countries.
Many of these jobs are things that improve quality of life for everyone else, too, like street beautification programmes. They do this alongside coming up with robotic means to also sweep streets, but yes, they find ways to hire people to just "make this place look nicer".
Chinese State Council: 2026 Employment Priority Strategy
http://big5.www.gov.cn/gate/big5/www.gov.cn/zhengce/content/...
Is it really that crazy though? Or is it an acknowledgement that the people in power would think like Peter Thiel and if given the choice between ending capitalism or ending humanity, would pick ending humanity?
Isn't the end of the humanity also the end of capitalism?
Not necessarily! In "Saturn's Children" by Stross, humanity managed to wipe itself out, but post-humans (evolved from former, ahem, entertainment androids) are doing just fine across the Solar System. There's a plot point where one of them discovers that some corporation back on Earth inherited full property rights to her body due to an old pre-human-extinction EULA, and she'll be recycled for parts if she ever sets foot in their jurisdiction (I'm paraphrasing from memory).
Do you expect competing AIs to give up on private ownership of the means of production and using them for profit?
And vice versa.
It'll turn into some starwars robot slave trading, Capitalism will be fine!
Explain a world in which capitalism ends without a fight?
Who said it needed to happen without a fight?
All the rich people become bankrupt or otherwise loose their political capital, new politicians are elected in that owe more to workers and thus enact anti-capitalist policy to win over the loyalty of workers.
That or, as rich people loose more and more of their political capital, workers get empowered and use direct action (such as strikes) to demand the end of more and more capitalistic policies, until there are none left. The rich get along with it because the alternative is loose everything.
Or capitalism ends in a big messy fight.
The world where humanity ends without a fight due to capitalism "winning"...
It will happen. Communism ended without a fight.
Edit: And today's industrial society is quite far from being capitalist. Feudalism is more close to reality, seeing as the one most important factor for almost everything in life is whether you own real estate or not.
That's a very loose definition of feudalism. However if you take in account nepotism and bribery, which are more and more widespread around the world, maybe we get close to that exchange of land for loyalty that was the basis of feudalism.
> HN: A site full of bold futurists where it's easier to imagine the end of the humanity than the end of capitalism.
That's an underrated comment! Totally agree. So so much of discussion on this topic sort of assumes business as usual (both literally and figuratively) and that AI quickly diffuses through the economy, sucking up all economically useful activity, with humans falling by the wayside as inevitable collateral damage.
There's a failure of imagination taking place here that there's no other outcome. Capitalism is an evolved solution to how human community works, but is not inevitable.
Massive changes in technology pretty much always involve wars and a reshuffling of power. This doesn't have much to do with capitalism at all. It's more of a question of how are those that currently have power going to adjust. In the case of technology we quite often see this end up with a concentration of power.
How to maintain stability in a world filled with greedy people in power that don't want to give up a single penny is a good question to ask at this point.
See also: Russia, 1917
> HN: A site full of bold futurists where it's easier to imagine the end of the humanity than the end of capitalism
HN is literally owned and operated by venture capitalists why are you surprised by this.
I guess op wasn't surprised, just called out for those who should maybe consider this angle of the situation also.
Unsurprising stuff can still suck, and we can still complaint about it.
If you're not economically useful, what pressure would you be able to bring against the government and it's drone-based police force if it does things you don't like? General strikes don't work in a world with no human labor needed.
Iran recently shot thousands to tens of thousands of its own people when they protested. If a sociopath were in power, what would stop them?
What's your proposal for ending capitalism in our working lifetimes?
>There’s no quick answers, but intuitively I think we all get that the % of jobs not doable by AI and robotics approaches 0 over time.
Can we really say that? I think that might be a prediction based on the current situation and on the assumption that jobs will neither be transformed nor created anew. We don’t know what the future holds for us in terms of jobs – or whether the problem of jobs will be a solved issue. Just as the blacksmith never gave a thought to what a mechanic actually was until he saw a car for the first time.
It is clear that at the limit, machines will be more capable in nearly every way. I suppose there might be some room for a few exceptions like working around big electromagnets that machines can't get close to... Does that sound great? Is there something else you had in mind which would be reserved for humans and fulfilling?
> I think we all get that the % of jobs not doable by AI and robotics approaches 0 over time.
I also think that will be the case. However, not all the jobs are worth being automated. Say, if you are a modern-day horse-riding instructor, I very much doubt the robots will come knocking the door to take your job. Not because they can't, but because the headache is not economically worth it.
It'd be uneconomical today but I don't see any reason to believe it will always be that way. Any time you want a high standard of living for people it creates an incentive to eventually make a cheaper way of doing what those people make their high standard of living with.
About the only things I can think to guarantee will have some constancy are "for the sake of being human" type things. Much like buying handmade bowls is something you can do out of pure preference/taste/ideology even though it's cheaper to buy massed produced ones.
[flagged]
I can see that going away too. Eventually the average horse rider is going to be someone of stature of a saudi prince at the rate horse ownership coupled with inequality is going. They'd probably be all about buying the Gemini horse teacher over learning from some lesser birth stableboy. I bet eventually all tutoring from a human is seen as some lesser ick by the rich. People already think as much about you today when you talk about say not relying on claude code. It is all about optics and virtue signalling pedigrees at the end of the day.
In the long run, we’re all dead anyway. (Unless something drastic changes.) That means how long it will take for robots to be practical matters. We don’t know how long it will be, but driverless cars have been coming for 15 years now and they’re still unavailable in most places.
People are more adaptable than horses and tech solutions usually require people to do some of the work. Dishwashers need loading and unloading. My robot vacuum still needs maintenance. Self-checkout means you do the scanning.
Amazon wants to get into drone delivery and there are cases where that might work, but the way news articles describe it, it’s not going to get a package onto my porch.
You seem to be living under a rock. Yes driverless cars started in 2009, but in 2026 it is very much here. Waymo recently crossed a milestone of more 500k trips I think, we will have millions of driverless cars in the coming years (we’re at 10k right now)
Waymo works in a single digit number of cities with extremely rare, highly favorable climates, and still operate only on fixed pre-mapped routes. And it's taken 17 years to get that far.
We are still a long way from a self driving car being the default type of car
Waymo still doesn't serve my city. We get too much snow/ice in winter and so they are slow to expand here.
There's always exceptions. To not beat a dead horse(hah!) in Texas you can still find people on horseback in the drive through line at Starbucks from time to time.
The cities Waymo serves are the exception even in the US, and worldwide it is even more an exception.
I give them til 2030 for that to become false in the US, maybe 2040 for the entire rest of the world.
For a while waymo didn't even serve the hillier parts of where they were deployed. These roads are challenging conditions for even a human rideshare driver, when you are down to about a cars width of winding road, with a moving truck trying to pass in front of you, and a dump truck trying to move behind you, cars increasingly piling on either end, no way to turn around or back out, everyone increasingly honking and cussing as if that would lubricate things. It is kind of hilarious in a way seeing how quickly things effectively blow up in those street conditions with just the perfect storm of two slightly larger vehicles passing at once.
Now you can imagine a waymo in this mix how it would just be on its knees in a situation like that. How is a waymo to interpret "fucking move already you dumb fuck" into any sort of meaningful instruction?
I'm reminded of that story about activists in SF requesting a dozen waymos at once at some one way street and creating a localized denial of service attack essentially.
But the horses themselves got the better end of the deal - they spend all their time in the field eating hay rather than pulling carts.
The phrase "sent to the glue factory" comes from this era where many horses made redundant were sent to be slaughtered for meat or raw materials.
No. The horses were made redundant and no longer exist in such quantities
So, uh, maybe don't dwell on that too much.
Actually, dwell on that a lot. It's probably the only instructive example.
The rich and powerful only suffer us because they need us. They cannot assemble every lead painted doodad by their own hands. If they could, they never would have hired anyone. So what happens when they no longer need as many people?
Society is already far too constructed around catching some of the wealth that just spills off of the people who own it, by being in their orbit. That activity is not actually productive, but since the only people with money are the ones that have a lot of it (K shaped economy), there's just not enough wealth left in the bottom part of the market to support anything else.
> The rich and powerful only suffer us because they need us.
Well that and there are more guns than people and if a million people come for you then there is no army that will be able to save you.
Looking at the news in the US, it feels like the wealthy and powerful are working that problem.
So long as they can keep us fighting amongst ourselves in meaningless culture wars. If there is ever a broad alignment of populists spanning left to right, it could get spicy.
You'd be surprised how easy it is to start a war, engineer water shortages and famines.
There are more horses today than ever... Oxen got the short end of the deal - almost nobody uses oxen even though historically they were the backbone labor of the farm, while the horse was for the right.
> There are more horses today than ever...
The global horse population peaked in the 1910s.
Apparently there are around 60M horses currently, while there's no world wide data for 1910.
Horse ownership isn't what it used to be, but the horse population seems fine.
The only data I can find ignores the majority of horses today - recreational riders. If you can find a better citation I'd stand corrected, but everything I can find counts race horses and horses used for farming.
Or on the shelves of Tesco.
Intuitively I think we all understand that the % of jobs not doable by steam and iron approaches 0 over time.
Give us a single example of a job that you believe can never be done by robots or AI.
reality tv star on something like real housewives. I think AI/robotics would have a hard time being that irrational and illogical consistently. That behavior is pretty unique to humans.
I feel like Neuro-sama is essentially already that, but for the Twitch / YT crowd instead of television. It's trivial to train a model to maximize engagement (as one can see gestures broadly), so I'm sure if we can get robots looking nice and human, we'll have Futurama's All My Circuits in no time.
Though I guess there was a token human on that show, so maybe you're right!
Waiter at a fancy restaurant.
The last job is going to be hairdresser, because billionaires have wives (not to assume all billionaires are men), and those wives will want a human to gossip to.
Amazing. I bet this can really help thousands of programmers who are simply not needed, because companies decided to replace them with AI agents. These philosophical ponders are very comfortable to do when your ass is safe. Most of people don’t have this privilege.
> what if what we loved about this job was never only the code?
It never was the code. The code is how you solve problems, make stuff happen. The code is a tool. It’s people’s brain that is of value. People learn how to use these tools the best they can. And now they are told that they should become mere supervisors for the AI agents. Simply by trusting the output and QA-ing the result. Because no human can keep track of the AI’s generated mess.
Yes, but OH, the destruction along the way.
My son wants to enter the software profession. He isn't just immitating his father, he really considered different careers and really likes this one. He is capable -- had nearly a year of experience working at a tiny company building a high performance database in C before they had to shrink; not quite enough experience to qualify for "experienced" jobs. He is even comfortable with using AI to build code.
But he simply cannot find work. No one is hiring entry level software developers (especially not those who are better at designing code than at networking colleagues). It appears that he will be driven out of the field because of the disruption AI is bringing, and I morn for that.
I don’t particularly like the horse to engine angle because it doesn’t capture the issue well.
When we automated physical activity people that were crowded out moved into knowledge work. Now we’re automating knowledge work so people move into ???
Leisure? Become a useless class? UBI? Something else entirely?
I agree with this too and I find this article extremely depressing.
The AI "revolution" as described here is the wholesale outsourcing of vast swaths of thinking and decision-making to a single entity. A project coded with Claude is a project coded by Anthropic. A business whose decisions are made with ChatGPT is a business run by OpenAI.
Is that what these people really want? To become a conduit through which two or three companies get to exert control over everything in the world?
The knowledge economy was already on the way out in favour of the performance economy even before LLMs. LLMs have also shown that they can produce in the performance space, of course, so it remains to be seen how valuable humans are in that position but so far it would appear that many are willing to pay for a "human connection".
Plenty if not all people put more value into handmade things, but they can't afford to have everything handmade. An IKEA coffee table is $15. Handmade one from a woodworker is $400 - $1000.
Absolutely. Hence why manufacturing died long ago, and why the knowledge economy was already on the verge of death even before LLMs, and why we were already moving to the performance economy: because in the performance economy the consumer and the buyer are able to be disparate entities, where the buyers can have deep pockets even if the consumers don't.
In the performance economy it doesn't matter how rich the consumer is, but what the consumer wants to consume does matter. Because the buyer and consumer are disjointed in the performance economy, the consumer isn't swayed by cost. However, if the consumer prefers AI performances, which they quite easily could, then humans are in trouble. But so far it seems human performances remain what the consumer wants to consume.
(Technically manufacturing and knowledge work could have established the same separation, but since the bright minds went to performance nobody thought to before it was too late)
How are your plumbing skills?
You mean my meat proxy skills while wearing a Ratatouille machine-vision head camera that tells me what wrench to pick up and how to crimp the Pex rings?
You're telling me Google Gemini robot 2 tying a trash bag isn't also gonna be able to crimp PEX rings?
It will. But having watched the video, I can tie one faster and cheaper. This week. Next week, who knows?
The solution is probably a trash bag with a mini forklift compatible lift rings, an optical center mark, a ratcheting closure, and make a raspberry PI do the computing millions of dollars of tooling to make a robot compatible trash bag but it might be worth it, deploy the bean counters!
Plumbing is kind of a hard job, definitely not fun for anyone above middle age. And given it will pay minimum wage, I think I'd just take a life of leisure instead.
Minimum wage? Where I live, the plumber median wage is pretty much exactly the national median.
Sure, but that is today. When AI decimates the knowledge fields, a whole lot of people are going to be looking at different career options. There will not be any shortage of people hoping to make money in the trades, and with that surge in supply will come a collapse in prices.
> There will not be any shortage of people hoping to make money in the trades, and with that surge in supply will come a collapse in prices.
And the worst case downward spiral will involve not only a surge in supply but a drastic drop in demand to go along with it.
All of those previous "knowledge workers" scurrying for the trades not only become more labor supply, but without their cushy jobs they will no longer be able to afford to hire trades people to prop up the demand side.
Double whammy.
Ah, then I understand you. But please, let's stop with this inevitability. If AI decimates, not when.
Apologies, I'm with you. I was speaking to the hypothetical, not trying to suggest it is inevitable. I think it is uncertain how this plays out, but I am an optimist and I think it ends up a net positive.
Why would it not? The economic incentives make it inevitable. Might as well expect water to not flow downhill.
It is very plausible that LLMs never get to 100% replacement for knowledge workers and the winners will the people who use both, leaving behind the people who thought all they needed to do was cut costs by doing existing workloads without humans. It remains to be seen if the constraint on quantity of knowledge work was the people or the work.
It remains to be seen, but a reasonable argument against this is found in the progression of chess AI engines. After Kasparov was beaten there was still a period where the best humans working with machines were still better than the machines on their own. Eventually (around 2020 I think?) this was no longer the case, having even the best humans on Earth in the loop was a net negative. There are reasonable arguments against this, but I've found them mostly unconvincing.
> Why would it not? The economic incentives make it inevitable.
Citation needed. Again, I wanna emphasize that I'm not saying it won't happen. I'm just saying that it's not a given that it will.
There is very little sign that current models can do anything but (at best!) augment most knowledge workers (with some exceptions). And there are very reasonable open questions about whether it's economically, and ecologically, sustainable to continue development at anything near the current pace. There is, in fact, plenty of indicators that even the current SoTA is the result of a financial overexertion.
Have you heard about "physical AI" ?
See my response to the similar reply at the top of the thread
Actually had a plumber here earlier today which was an ordeal so have been asking that myself too...
> Become a useless class?
Like our capital class overlords?
I’ve been coding for over 20 years. I may not be the perfect coder, but for me, the real goal has always been solving problems and coding is just a side effect. If I could solve an operational problem with Excel, I would! Code, once created, is a liability I have to carry.
So I’m embracing AI-assisted coding with open arms. It’s not that I don’t enjoy coding. it’s that this approach lets me solve problems faster than ever before.
Ok but you do realize that AI can also solve problems? The thing is it's not just taking away "the coding" it's also taking away the problem solving (of course I'm not talking about all problem solving).
That's the real risk here is that we are getting to the point (probably already there for a lot of stuff) where you can just point it to your already existing codebase and tell it your problem and it can basically do it.. I mean I've tested that in practice, sometimes I'll have a clear direction of where I want to go or what I would do.. but I don't really give it any hints of how to solve the problem, and many times it comes up with a similar result or at least one that also works completely fine.
So I think part of the issue is that people like you and me I feel like are going to be so much less needed because an AI can just do it instantly for basically free.
Anyway, I know this is a slight hyperbole.. but it's also not. It kind of sucks to realize that a simple prompt can replicate your work 95% of the time and do it basically instantly.
> where you can just point it to your already existing codebase and tell it your problem and it can basically do it
Who is deciding the action/s to take in response to the problem? How do you know which is the right action to take?
"A computer can never be held accountable, therefore a computer must never make a management decision."
Currently we have C-level people saying "I want a website that sells widgets". Then product managers make a plan for all the pages/functionality on that site. Designers style the site. Managers divide up the work into smaller tasks. Tech leads steer a group of software engineers, and software engineers do the actual implementation.
2 years ago AI was a novelty and essentially just "fancy autocomplete". Today a skilled software engineer can use it to mostly one-shot the whole website with a detailed technical prompt and then iterate on it from there.
What happens 2 years from now? If progress continues, I expect we'll start walking up that chain of command. First you eliminate the software engineers and just have the tech leads do the prompting. Then you eliminate the tech leads and have the managers do the prompting. Then you eliminate the managers and designers and eventually you end up with star-trek-esque programming where some C-level person is just telling their computer "I want a website that sells widgets. Good! Now make the navigation bar blue. Could you make the font bigger?"
So what are your plans for when the wealthy no longer need you?
I look at this and I feel the reverse: what do the wealthy bring to the table now? previously we had to work for them cause anything worth doing that could become a business needed a bigger team. now, a developer doesn’t need anybody above him to make stuff, and can potentially work on doing things they find valuable which imo something we always strived to but couldn’t afford to. in my mind, it’s really about ideas and execution now. and execution i bet a developer who wants to make something can do better than anybody higher up the chain for the forseeable future.
sure there’s things you still need investment and a physical presence for. but i reckon there is money and careers to be made in purely software based businesses still, and i think that value will mostly be captured by individuals and small teams moving forward.
The wealthy bring the capital required. The problem is fundamentally the commoditization of intellectual labor. It will simply be a game of whoever spends the most wins.
even in software historically, larger teams/more money have never produced better software. sure, not disagreeing that may be they’ll have all the best LLMs and the hardware to use them for a while. But I don’t foresee open source weights to the public to lag that far behind, and hardware that can run them to become more affordable.
you know who _own_ the means of production more than ever? the rich. they control your labour environment. they control politics. so you're the opposite of independent from them.
i could setup a local llm that’s nearly frontier level if i wanted to. i doubt i’ll ever need to though since tokens are becoming so cheap. although i am well off, i am not wealthy. and i only foresee this becoming more affordable in the future. may be i am wrong but imo an era when capital is less important is already here for the purposes of what i described.
my point was that for anyone who wanted to make software they couldn’t due to lack of free time or funds, they likely can give that a go now.
Then I will find other problems to solve. Do you think AI will solve every problems for humanity? it has nothing to do with coding as I mentioned earlier.
Just because there are open problems AI cannot solve, it does not mean you can make meaningful contributions. AI might become more capable than you in every way. Say AI cannot solve P vs NP. You can dedicate your life to the problem and will not make progress. Similar for more worldly problems like new treatments. Or societal problems. You are the cow.
What kinds of problems do you think someone will pay you to solve after AI advances for another 10 years?
The internet enabled the massive complex systems we have today (public cloud being the largest in scope probably, but there are sprawling libraries like everything around JavaScript, modern web browsers themselves maybe).
Without the internet these systems can not be maintained, we need all information about them available instantly. We also need people sharing how to solve problems at a global scale which the internet enables.
Personally I think LLMs will just allow even more complexity. This will create more situations that need human intervention to solve problems. And also new problems we have never seen before.
Hah I have no idea. If we reach AGI in 10 years we will still have problems. Otherwise we all would be enlightened by then, right? Until then I think mental health would still be a problem. Maybe I should get a therapy license now.
> Then you eliminate the managers and designers and eventually you end up with star-trek-esque programming where some C-level person is just telling their computer "I want a website that sells widgets. Good! Now make the navigation bar blue. Could you make the font bigger?"
Those C-level people won't want to do that. Plus in a competitive economy everyone else can do the same so there will be competition to remain relevant.
1 website won't be enough, you'll need 10 and each site needs 10 times the features or your buyers will look elsewhere.
At the end of the day the C-level person delegates to a techy person, it's not the C suites job to care about font sizes or navigation bar colours.
> Ok but you do realize that AI can also solve problems?
Why have companies with infinite amounts of AI and expertise not deployed comprehensive end to end solutions for business problems?
Every solution they've come up with is "here's a tool. The solution to make money from it is left up to the reader". Why would they leave all that money on the table?
What you said is valid but I think you are still constrained in thinking that the code is the solution, though. I wish I could word it a better way. Maybe we need a word calculator /s
It's not right a lot of the time (yes, even the latest models) based on the problem you're trying to solve for your users. (That's one thing we can keep doing).
It doesn't always know the right thing to fit the need, so you can now figure out what that looks like and design it appropriately (and try 10 options in production and see which one wins!)
But also you can just do a lot more work. If the prompt can replicate 95% of your work, now you can do keep moving forward faster and do a lot more work.
This does mean, of course, that there will be less of a need for software engineers that are only capable of reasoning about the edges of applications or one narrow part of the whole discipline will get squeezed out. If a computer can do the job, it's now less valuable of a skillset than it used to be in the market, and that hits wages hard if your only contribution to the market was "I know how to create React components when someone tells me what is in them, and how they need to interact with other components"
Happily, despite the amazing productivity of AI, there is still so much absolute CRAP software out there, and so many different places we can apply better options. Bureaucracy and poor management will continue to create new opportunities.
My grandad ran a small letterpress shop and when phototypesetting came in he didn't retrain, he just did smaller and smaller jobs until it was wedding invitations.
“Retraining” is mostly a myth. Former skilled workers just downgrade to service jobs, pushing out younger, less skilled workers in the process.
The NYT touched on this issue in a podcast about NAFTA: https://www.nytimes.com/2024/10/08/podcasts/the-daily/americ.... AI is going to be NAFTA on steroids. Politicians will be able to point to growing top-like GDP to say it was a good thing. But it’s going to brutally hit a lot of people and most of those folks will never regain a secure economic footing.
Some skilled workers retrained. Many of them refused to learn.
Don't be the old dog who can't learn new tricks.
Retraining for typesetters meant becoming "prepress" guys, and going from the wide array of tools and hand skills they used to need to Photoshop, Illustrator, and digital imposers and plates. In the process, the job went from a middle-class job to a minimum-wage job, and the bosses who used to respect you (after the union that used to protect you disappeared) became fly-by-night scumbags who treated you like dirt.
As far as I can tell, the entire business for someone who was good at running a press is fancy wedding invitations and business cards with raised type. I am speaking as someone who waltzed into (and later out of) prepress because I knew the digital tools, made no money, and talked with those guys a lot about the past. This discussion would be while we were printing wedding invitations and fancy business cards.
It's simply a different, low-status job now. Retraining is no different than training the first time, because if you want to make a living you either have to somehow be better than the people who started with the new tools (and the new tools don't require as many people) and maybe get into printing packaging or manuals, or you need to do something completely different, and compete with people just starting their careers.
Going to college again is not a trivial thing. The mortgage will still be due.
Right, retraining often means a major switch. I don't want to go back to college either...
Most of the time, you just have to learn what makes the new trick the same as the old trick.
The last hundred times we automated things, we automated muscles. Today, the number of people employed purely for their muscle is very small.
This time is the first time we've automated higher level reasoning.
I'm not sure I can quite agree with that. I'd say that spreadsheets definitely reduced the need to manually calculate things, and calculators before that, and the abacus before that.
I basically agree with your "higher level reasoning", but not your statement that we previously only automated muscles.
see: when 'computer' was a job title
Also relevant: https://en.wikipedia.org/wiki/List_of_obsolete_occupations
> Today, the number of people employed purely for their muscle is very small.
I also believe we underestimate how much that costs us too. Yes, it is good that fewer people are breaking their bodies on brutal physical jobs.
But there are many people in the world for whom their physical strength and dexterity are perhaps their most valuable asset. With the loss of those jobs came a loss of an avenue of meaning for them.
I could be wrong, but I have a hunch that the whole manosphere alt-right conservative thing going on in the US is driven in part by a generation of young men who don't find intellectual work to be appealing but don't see many alternatives out there either. It's telling how closely Gamergate is tangled up in that world and how many games are basically simulations of heavily physical prestigious roles.
A recent Onion headline really hits the nail on the head: "'This Here Is Probably Our Bestselling Love Seat,' Says Man Who Would Have Been Powerful, Revered Warrior 4,000 Years Ago"
I think a more obvious effect is that now everyone is obese.
Compilers automated higher level reasoning, at least to a small degree. LLMs themselves are basically just compilers with a much, much, much, much more expansive set of built-in language constructs.
(No, compilers are not deterministic)
Can you describe the last time you asked your compiler how you should best approach an interpersonal problem, to research an investment thesis, or to figure out an org chart by analyzing communication patterns within teams?
People aren't just using AI for writing code, but for all of their decision making. AI is far from limited to writing code.
> Can you describe the last time you asked your compiler how you should best approach an interpersonal problem
Sure. I wrote out specification that described the problem. It compiled that into a description of how to deal with it.
> People aren't just using AI for writing code
Compilers have never been about producing code in particular, though, even if that is where they are most commonly used. A LaTeX compiler, for example, outputs books. To think they are about code exclusively is a grave misunderstanding of what a compiler is.
> Compilers have never been about producing code in particular, though, even if that is where they are most commonly used. A LaTeX compiler, for example, outputs books. To think they are about code exclusively is a grave misunderstanding of what a compiler is.
And what, pray tell, are those books output by TeX? That's right, PDF or PostScript code.
> Sure. I wrote out specification that described the problem. It compiled that into a description of how to deal with it.
I might just not know enough compilers, but didn't you write the specification of how to approach the problem, not just the problem itself?
I don't think I knew how to approach the problem. There would have been no benefit in using the compiler if I did. Kind of like, if we want to talk about code, many coders don't know how to approach making a hashmap. They don't need to. The compiler figures it out for them, as it did me for my problems. I'd call the input I gave a specification of the problem, but words are in the eye of the beholder, I suppose.
If we take "write the specification of how to approach the problem" at literal value, you might be narrowly thinking of a low-level compiler, or what sometimes gets called a transpiler? However, compilers encompass a much wider spectrum than that. Generally they enable users to operate at higher levels of abstraction.
I think you're talking about more than just a compiler, though standard libraries are often distributed with the compiler. Take Rust for example, it's not like the compiler has special integration with hash maps, it's just that the standard library has a standard implementation and the compiler lets you refer to that implementation without having to repeat it yourself. In a way you're just including parts of another spec in yours. Specifications do this all the time, for example by using English. You don't have to define all the levels of abstraction, only the ones that matter.
Even so, I'd argue that you still have to define how to use those abstractions rather than only the result you'd like to achieve with them. Sure, maybe the abstraction has a ".achieve_the_result()" method, and calling that method is how you achieve it, but often there's a bunch of stuff you fit together to achieve a solution to your problem and the compiler doesn't necessarily do that for you.
I think there actually may be some compilers that get closer to this, for example some problems can be expressed quite elegantly in Haskell or APL and the compiler will do a lot of black magic behind the scenes to get you a performant executable (somehow). But for most traditional compilers you are still going to have to define how to approach your problem, even if only at a higher level thanks to libraries.
> Sure. I wrote out specification that described the problem.
The LLM usage I see from most people has the LLM write out the specification, asking clarifying questions as needed. If you're using LLMs as you describe, you're using them ineffectively.
> The LLM usage I see from most people has the LLM write out the specification
If the compiler is writing the specification, what purpose does the person serve? That is not what I'm seeing out in the wild, though. Perhaps what is happening in research labs aside, compilers, even those implemented as an LLM, don't do anything until the user specifies their intent.
In fact, you literally recognized that in your earlier comment, asking about my action to specify my problem. Not sure why the story is changing now? There is probably a day where the systems solve your problems before you even know you have them, but we're not there yet. The technology hasn't advanced that much in the span of a couple of hours.
> asking clarifying questions as needed.
Sure. Even the earliest, most primitive compilers would seek clarification if there was something that it felt required more information. That's expected.
> If you're using LLMs as you describe, you're using them ineffectively.
LLMs are just compilers so they can't do anything else. Maybe you're thinking of some technology that is layered on top of a compiler like an IDE, REPL, or something to that effect?
> Sure. Even the earliest, most primitive compilers would seek clarification if there was something that it felt required more information. That's expected.
Can you tell me what the last question GCC asked you was, and can you rate its coherence in comparison to your colleagues questions?
> (No, compilers are not deterministic)
Having used, designed, and written compilers, this is news to me. Please, elaborate.
tldr: Race conditions when building on multi-cpu systems (which is all of them these days), and timestamps sneaking in everywhere.
Oh, no. Timestamps. Truly the sign of higher level reasoning.
This is a tired, idiotic analogy. For one, if LLMs are compilers, they're terrible compilers because they can't be reasoned about the way one wants to reason about a compiler (but I assume you have a weird take here since you claim compilers are non-deterministic).
But more importantly, if you insist on treating LLMs like compilers, then the source language – natural prose – is the worst programming language ever invented. Coding the world in brainfuck is a better idea than where you lot is suggesting we go.
> This time is the first time we've automated higher level reasoning.
That’s very generous of you to call it that way.
It’s a parody at best.
That's an optimistic view -- and even if it is, as the cryptographers like to say, "attacks only get better".
My key thought is this, many of us necessarily work in low-purpose, boring corporate development jobs. The code was the only thing making it enjoyable, because problem solving was the skill we enjoy exercising and you get that in spades with programming, even if the actual business is soul sucking. This is the reality for many professions not just software. The second part of that reality is that we have to work to pay the bills, and stability is an important pillar that many families rely on. I mention that to answer the inevitable question "why would you work there then?". There isn't an abundance of fulfilling purposeful jobs unfortunately.
That problem solving aspect is being replaced by AI, and that means anyone in a position like that is losing the piece of the job that they enjoyed. The more apt analogy is that the farrier worked in a farrier factory, and the horses were actually mass produced statue horses with no purpose but making money. He still enjoyed working with his hands, but now they have replaced that part of his job with a button to press. Now he presses the button in the horse statue factory. He still has a job but now no joy, and eventually they will automate the button too. The outlook IS bleak, and many of us will move on to totally new industries and that's fine. Some will stay because they have a rare job where they have agency to problem solve on the layer above code. Many still just press the button to feed their family.
I agree with your first paragraph, a majority of corporate programming jobs probably suck. However I'm not convinced that AI will reduce demand for these positions to in a major way. It will instead change the required workflow. I've got a couple of industrial revolution points I'd like to share; The Jevons Paradox addresses demand concerns. Software is now theoretically faster and cheaper. More projects are now financially viable, since total unit costs are down! The other side is the Automation Paradox, where now even more specialised knowledge is required to fix and maintain AI systems.
Your disenfranchisement is valid, however fortunately (or unfortunately?) do-nothing soulless corporate programming jobs have a good chance of sticking around, if in a changed way.
I think this is true, but I also think that AI has made the mental burden of putting up with a less than ideal job much lighter. I am no longer mentally exhausted at the end of a long working day, yet I have made a meaningful difference in terms of technical progress. And, at the end of the day, I am still bursting with energy about exciting personal projects I can now take on and make real progress on because I'm not dead at the end of the day.
Not sure if this fits, but the mechanic, rather than performing hard manual labor mounting horse shoes on horses, now just has to read software error codes during the day, and gets to enjoy driving his sports car in the evening.
Back in the '70s my dad used to tell a joke about his dad. He said that my grandfather had a job that supported his whole large family while my dad was growing up. But unfortunately technology caught up with him, and he was replaced by a device just this big, and he held his fingers apart to demonstrate.
Then he said the real disaster hit: "And then my mom got one."
Grandma was there when he told it more than once, and thought it was hilarious.
It usually came with another joke about the time he was caught with his pecker in the pickle slicer. He was fired, and so was the pickle slicer.
In the seventies you were expected to tell jokes like that and allowed to laugh at them.
I’m trying to figure out this joke. Is the punchline that he was a gigolo and he got replaced be sextoys?
I think the punchline is that workplace romance usually gets both parties fired, regardless of who initiated. Hence, doing "things" with the pickle slicer means the pickle slicer gets fired for participating...
They're both misdirection jokes.
The first one sounds like the dad was replaced by computers. But then reveals the dad is a sex worker and being replaced by sex toys.
The second one sounds like the person was being intimate with dangerous kitchen equipment, then reveals that a "pickle slicer" is a job for a human and not a machine.
That was my interpretation but nobody ever explained it.
It may be funny when it's a joke, not so when it's real.
My grandpa was a shoemaker. The real change in technology left his family in poverty in Chicago during the depression. I didn't hear anyone joke about that, directly.
Jewish humor is frequently about laughing at your own pain. There's a genre of Jewish jokes created in concentration camps.
Well written, and good to reflect on.
However, the one thing I feel like is missed by some in these conversations is that what's being built is a general purpose thinking machine, not a coding machine. How are we going to transition from writing code to solving problems when the machines are solving the problems?
You can't outsource understanding, and to the extent LLMs can understand anything they can only learn through training. So until LLMs can truly learn, rather than bringing themselves up to speed with huge contexts, you'll probably still want some people who understand what it's doing.
I think Navier Stokes is a brilliant case study for all that it didn't do. An AI 'solved' a famous problem, of a kind that is impossible to cheat, yet the reaction is incredibly apathetic. Why? Because it apparently failed to do the thing people were hoping it would, but what exactly? Of course it also manages to fumble the delivery by antagonising the people most interested in the result, but still.
The honest AI boosters do claim we won’t solve problems anymore, we will live animals in a zoo, overseen by machines of loving grace.
The retard AI boosters, tell us to be AI first, learn this harness bro, loops are yesterday, now it’s graphs, use 100 agents a minimum or be left behind, actually world will require 100X more software engineers.
> we will live animals in a zoo
Why would we still be alive? Once we have advanced enough AI inside not only computers but robots, the rich will no longer need us. They will have the robots farm the land, clean their homes, fly their planes, etc. We will then have no income and therefore no way to feed ourselves. Most people do not own enough farm land to keep themselves alive.
In the past, such imbalances of power were resolved because the rich still relied upon the poor for everything (security, food, etc). I think we're approaching the point where even armed insurrection is no longer an option with the advances in automated killing machines. We're not there yet, but it's coming.
Other than Uber and Uber eats, I can’t imagine myself doing anything other than those. I might get a security guard certificate and guard empty halls during the night, which quite suits me.
Not a handyman, not suitable for physical jobs, I think I’m cooked! This is the fear that is always on my head.
What makes you think those jobs will still be available?
Uber : Getting replaced by Waymo
Uber Eats : They've been experimenting with ways to replace humans for this. I think the most likely scenario isn't those silly robots on the sidewalks but rather just a Waymo. Have the restaurant send someone to carry food from the restaurant to the Waymo idling at the curb, and have the customer come out to the street to pick up their food from the Waymo. We lose the door-to-door convenience but I think a lot of people would take that trade to eliminate the cost of tips and paying the delivery person.
Security Guard : AI monitoring security cameras and automating calls to police sounds trivial. There are also already "security robots" that move around so they could be dispatched for scenarios which do not warrant a call to the police.
Handyman / physical jobs : While I think these will last longer than the others, I think they won't last much longer. We've already invented all the necessary hardware. The limiting factor was the software. What happens when you take an LLM from the year 2030 and give it a pile of tokens to write the software to drive the robots to do plumbing/building/whatever.
> Uber Eats : They've been experimenting with ways to replace humans for this. I think the most likely scenario isn't those silly robots on the sidewalks but rather just a Waymo. [..] We lose the door-to-door convenience but I think a lot of people would take that trade to eliminate the cost of tips and paying the delivery person.
Amazon is already able to do deliveries for some packages by drone in some cities. This seems perfect for it.
I don’t think they are not going be impacted. So yeah, maybe I’ll be jobless.
This is my dilemma as well. I can't think of a role that won't be impacted so I don't know what to train for. When I was younger, my fallback was always going to be "truck driver" but that's going to be one of the first dominos to fall.
I have no idea either. And I’m in my mid 40s which is a pretty hard place.
If I’m younger I’d probably find some trade and do it. I think it’s safe for another two decades maybe. And I wouldn’t bore a child.
If I’m older I’ll just call it the end and retire early. My wife still works and I can live at around 300-500 CAD per month for all expenditures.
But now it’s a bit scary if I lose my job.
Yeah in Estonia we literally have robots deliver food (starship).
So naive.
Automation will take everything it can in a race to the bottom towards higher profits.
Your great great grandfather changed, and the automation didn't keep up.
This time it will, and there will be no safe harbor.
I don't think we can say what will happen, but I agree that history offers no guarantees for the future.
Significant unemployment leads to civil unrest, which leads to government intervention.
Also, it leads to significant lower purchase power, which may indirectly decrease profits, at some point ? ("Firing your customers" works up to a point)
You are not the farrier in this analogy. You are the horse.
No, he is the farmer all right:
https://manuel.darcemont.fr/projects/
We are the horses.
> something happened once in the past and turned ok hence you should never ever worry about other vaguely similar things ever again
That's how it sounds to me
Don't forget blatantly ignoring all the times something similar happened in the past and didn't turn out fine.
Sticking your fingers in your ears and singing "Lalalalalala" will in fact sometimes work for some people, and that is not a good argument for everyone to do it.
I like this article, but it took decades for cars to take over: we needed to build vehicles, build roads, build up gas production and distribution... as Bill Gates wrote, one of the differences with the AI disruption is how fast it's coming, since it's almost immaterial (except data centers of course). It may be too fast for humans to adapt like they did in the previous revolutions.
The current trend of comparing AI development to the development of motorised personal transport makes it very easy to remind everyone that undoing the effects of decades of blind "progress" once the negative externalities become unavoidable is much easier said than done.
Unfortunately, work from home made it easy for companies to realize how easily a chat window could fulfill employee interactions
And it threatens not a single profession, but all people doing intellectual work.
It can replace all work, IMO. I'm doing intellectual work that replaces manual labor with machinery and computer vision systems. It's massively sped up by AI.
The grandfather must also have survived two world wars. The transition from horse to cars happened within a broader context and was greased by blood.
I find it curious that this is often elided when discussing technological transition. Sure the whip makers went away but they probably also ended up dead in a trench.
So far the impact of LLMs have been very much overstated. Like most (all?) tools, they require a human in the loop to work as intended. This fear of being replaced by an LLM is unfounded and irrational.
The alternatives that may cause such mass displacement like World Models and such are yet to be proven.
LLM's went from generating incoherent garbage, to weird text that seems coherent but isn't (gpt-2), to what we have no with agents passing some of the hardest benchmarks, necessitating benchmark development just to be able to keep up with the rate of progress. They did this in about 5 years.
The sheer rate of progress is what is meaningful. If LLM's agents were frozen now without any further improvement, your position would be must more defensible. But that's not what's happening, we are getting breakthroughs almost on a weekly basis.
It feels to me like the advancements are not accelerating, though, like we're nearer the top of the S-curve than enthusiasts think we are. A lot of the recent advances seem as much a harness improvement as an actual model improvement. We've made exceptional progress in the last couple years, but spent an exceptional amount of money to do it. A moon shot, basically.
Honestly, my gut feeling (worth every bit of what you are paying for it...) is that until we get actual intelligence instead of a good facsimile, we're not looking at an existential problem or a UBI-level problem, even, just a big change in tooling.
Hey I hope you are right. I, like so many of my fellow engineers, am not exactly thrilled at the prospect of my role changing from underneath my feet and the uncertainty that comes with it.
We are making steady progress. It’s not exponential per se like we used to see but in my experience models are steadily getting better. Also breakthroughs can be sudden.
"So far" is a statement about the present and past. "This fear" is primarily about the future, though a few jobs already have been affected. So it's not like there is zero evidence.
You’ve been living under a rock dude, have you even tried using it for maths or coding. No human in loop needed.
While I appreciate the point being made here, a tangential thought:
> A century later, the torments felt by my great-great-grandfather are forgotten. The garage feels like it has always been in the family. My father loved his job like few people I know, as if it had run in the family forever.
Arguably, the internal combustion engine is one of the most unsustainable and devastating technologies invented by humans. The jury is still out on whether our species will survive the environmental impact wrought by its effects. So much benefit, but so much dependence.
Dont you think silicon based AI will soon outstrip the energy it needs to run at AGI scale? Since we can already grow brain organoids (and let them spread out flat instead of being crumpled up in a ball) we could scale brain size enormously. And since human intelligence only uses 25 watts, imagine the superhuman intelligence of a 25kW brain (the energy used by a server rack). Every neighborhood could have its own superAGI. And not only that, but new brains can learn from old ones (a better approach than distilling models)
I like this idea, but I don't think it's feasible. First, neurons are slow. They fire at most around 1000 times per second, while transistors switch billions of times per second. Second, although we don't fully understand how the brain computes, it doesn't appear to work like a discrete state machine or to rely on symbolic manipulation. We could probably make wetware emulate CPU-style behaviour, with discrete states and symbolic operations, but we'd likely end up worse off than with ordinary silicon.
25kw brain would probably smell like bbq
> Many of us, deep down, are driven more by the joy of making things, solving problems, and seeing people use what we built than by the act of coding itself.
I've definitely always enjoyed the end-result but only for the fact that it proved I have overcome challenges and slowly but steadily inched one step closer towards mastery. Both are now gone, since mastery is not required and challenges are absent except the soul draining exhaustion of reading markdown all day. Now I'm just producing outcomes for my boss, rather than my own fulfillment. The journey has become a farce, I will ride this out and make my money but I don't care about it much anymore.
That being said in my free time, as in Hobby I am doing pentesting manually for fun. Because productivity is irrelevant for my hobbies, so I can still keep inching towards mastery and overcome challenges, at least on days where my job leaves any trace of surplus energy at the end of the day.
Can you not seek mastery with the new tools, though?
It does feel a bit like building on quicksand, but coding has always been that way for me. Well, less so with embedded than webdev, but still.
I used to find joy in knowing how things worked inside-out. Whether it was Postgres, or browser engines, or some detail about compilers... I used to find joy in developing the little techniques that made programming NOT building on quicksand. There was satisfation in what you had done being robust.
Pardon my lack of creativity but I'm not even sure what mastery with the new tools (LLMs) would look like. When you are so abstracted from the details and don't have to know anything, what _is_ mastery? (I am not asking rhetorically and if people have overcome this thought I would be interested in hearing from them.)
Knowing how to ask a machine questions in the right way? Playing as a manager in control of non-human developers? It all sounds really silly and something that is difficult to aspire to (and I can't imagine explaining my job to others with a straight face). I think I find it even more egregious because it's all built on paying rent to the LLM operators.
It's not that I don't enjoy results in their own right sometimes but I think I am definitely in the 'this is giving me an identity crisis' camp.
How do LLMs work? Yes there's a model and there's inference, but dig deeper. How can you, today, with < $50, train a small GPT-2 class model, how do you fine tune it, what training data do you give it, how do you give it training data, what algorithms do you use for reinforcement learning, what even is RLHF? You don't have to know any of those details to use ChatGPT, just like you don't need to know how a D flip-flop works in a CPU to write some code that runs on one, but if you're looking for feeling like you know what's going on, at their heart, LLMs are still next word prediction engines, so knowing, deeply, how that engine is created and works, would be my idea of mastery of it. Karpathy has a good set of tutorials on this.
>How do LLMs work? Yes there's a model and there's inference, but dig deeper. How can you, today, with < $50, train a small GPT-2 class model, how do you fine tune it, what training data do you give it, how do you give it training data, what algorithms do you use for reinforcement learning, what even is RLHF? You don't have to know any of those details to use ChatGPT, just like you don't need to know how a D flip-flop works in a CPU to write some code that runs on one, but if you're looking for feeling like you know what's going on, at their heart, LLMs are still next word prediction engines, so knowing, deeply, how that engine is created and works, would be my idea of mastery of it. Karpathy has a good set of tutorials on this.
Knowing how an LLM is built, is trained, inferences, isn't "mastery" in the sense that crafts people use the word.
Mastery to a craftsmen is as much a product of mechanical performance as it is knowledge. For a software craftsmen, that means knowing how to code, debug, and architect a software application by hand.
This, plus since you can't realistically run these LLMs at home, the motivation to get invested in them is not really there (for me at least)
I mean, yes I do find the mathematical background of LLMs fascinating, but it's not going to be something that I can put to practical use (much), if that makes sense?
When I would code by hand I often found I was concerned with the house of cards in my head, am I remembering the right variable names, did I forget something that would lead to an exploit, etc. Now I can direct the LLM to do the work and the skill is split between having an understanding of the interactions of the technologies, choosing tech that will be less of a pain to update/migrate. You can even put some LLM-bow grease into finding more efficient approaches catching corner cases etc. And just by asking I can have test frameworks setup, I can drill into some pet peeve and it takes less time. Now we still have lots of ops moments, like the radicel distributed git forge forgetting to use encryption for private repos, and I run into this class of problem in my own casual LLM use, but that's where our taste and attention to detail powers need to be spent in this modern developer world.
Thanks for your perspective
I think one thing that troubles me is that I so far have gotten quite far by 'caring about the little things'.
Now we live in a world where this feels, for the most part, unsustainable to keep up with. When directing LLMs to code something up, it's still occasional useful to have the knowledge and experience of those little things earned over years, but at the same time 1) it seems like I run into fewer opportunities to add to that experience and 2) it seems like you can often just prompt LLMs to 'grill' a design a few times, it will probably surface some of the most important tradeoffs
Yes, the grilling part is where the LLM energy drain is for me, I haven't released any of the software I've been working on yet because they lack polish in my opinion. They may have nobel goals, but no point unleashing more slop on the world. Often when I ask for a correction something new breaks, or it reveals the next corner the model and I didn't think about. I've seen the models come a long way in a short time. I'm tinkering and paying by the token for deepseek flash picked for its economy and using it to teach myself and form opinions on languages and frameworks that are new to me. It's hard to describe the worth of these opinions, but I think they're important, kinda like why I read a lot of HN looking for morsels of sentiment around the tech industry as represented here.
It's funny though, llms are quite terrible at most things really.
They can build things, but it's all terrible really. Their "ideas" are terrible.
Well, I'm somewhat minded to agree with you, but they are fast enough, cheap enough and don't get bored of tedious work, which makes them passable at many things. Maybe.
I feel caught between two worlds, I know people both very 'for' and very 'against'. The business is very keen for people to use them to go faster.
For non-trivial changes, output so far does not meet my quality standard and I tend to spend a lot of time cleaning up after them, probably more than I would have spent doing the work myself. This being much more boring, I hasten to add.
Their comments suck so I feel like I have to go in and rewrite them all.
Maybe I should just let go of the quality standard and accept that in exchange for steaming ahead at full speed, but I'm reluctant (and not just for me, there are also coworkers who I feel like I would be letting down).
I've always been either entirely self-directed, choosing my own goals, or almost entirely self-directed, being given high-level goals, and getting to choose my own means to accomplish them. So my opinions are probably different from somebody who was a smaller part in a chain.
You say that you liked learning how things worked inside out, whether it was Postgres or whatever. You can still do that. You can learn it at a high level, and then it becomes like a magic word in your arsenal. You don't have to know or remember the details of implementation, but only the nature of its costs, benefits, and synergies with other systems. You just have to remember that high-level abstraction, maybe just a single vocabulary word, of a low-level reality - the exact syntax is really no longer a concern.
The better the agents get, the less there is playing manager. Really your job now is to have and execute vision. Execution is largely a function of having a map of the high level vision and low-level realities in your head, or just having in-the-moment intuition about these things... then just speaking it.
Also, there is developing personal libraries. You may find that in your business you repeatedly tackle similar problems. You develop and refine a library that your future agents can utilize. You develop prompts that make your work more automatic and more robust. And when I say you, I mean, you direct your agents to do it.
It's not so different from what it always was, except that years get compressed into months. What was slow and tedious is now... less so.
I don't know how much advantage I have in this new world, having programmed (sometimes more, sometimes less) for 30 years. My brother has never written a line of code in his life. He's been heavy into AI for a while, and he has developed some really incredible stuff. I think he may have already outpaced me. His $1500/mo+ in subscriptions definitely help.
Did a tailor gain mastery of the sweatshop machine? Did mastery of the automated sewing machines give the same joy to the tailor or was it mind numbing meaningless activity that he would be happier to be free of.
> It does feel a bit like building on quicksand
that's a good analogy and puts into words the way i've always felt about software development. Every so often the entire industry rolls over into something new. I've been here 30 years now and, in my experience, the skill that counts is being able learn new skills. To go back to the analogy, to build something new on top of what you've built in the past before everything is swallowed by the quicksand.
How can you say there are no challenges and things to master when working with AI? Unless you are just doing "Make this. LGTM. Now make this".
> "Make this. LGTM. Now make this".
I think well north of 90% of people in software engineering have job expectations that, when assisted/automated with AI, boil down to exactly that.
It's a pain in the ass to work at such a job and find challenges or things to master--your superiors don't want you to do that, there's not enough time in the day, and you're tired from LGTM-now-make-this work.
That's most software jobs, because that's what most software companies want.
Because I literally can just burn tokens instead of using any of my brain cells and the outcome is way better than what I would have achieved thinking deeply and intricately about the implementation because Opus 5.5 at xhigh implements way better than I ever could and I'm paid well and have studied for years and it's not a low-level environment at all.
Since opus 4.8 the tools have started surpassing me and now they are far gone, so implementing by hand would mean sacrificing quality and my main goal as an engineer has always been quality so even though I hate it, I must use the best tool for the job.
The Industrial Revolution took over 100 years. You could argue that WW1 and WW2 were direct, structural consequences of the Industrial Revolution.
This has taken months to up-end multiple industries.
I don't think the article makes a fair comparison.
The information revolution has been going on for >50 years. AI is one event in a broader arc. Wait 10 years when the humanoid robots become reasonably capable and affordable. We’re in the middle of it, but still so so far to go.
> Wait 10 years when the humanoid robots become reasonably capable and affordable
We've had humanoid robots that could walk for what, 30 years? While technology marches ever forward, and sometimes makes pretty good leaps, 10 years for both capable and affordable seems pretty optimistic.
I’d agree that the word “reasonably” is carrying a lot here. But call it 20 years, no question. Look how far computers came from 1970 to 1990, that’s probably about the pace to expect, or faster.
I don’t know, somehow the pace seems different, these transitions felt more like the last generation retiring, and the next generation doing something different, instead of it happening to everyone at the same time.
Yes, but this time, we're the horse.
Only if you accept that you are as stupid and unadaptable as a horse.
A less misanthropic alternative: The technology required to replace the person is much smarter and more adaptable than that which was needed to replace the horse. Because of this, you can end up in the horse's role in the story without having to be as dumb as a literal one so long as it's this other lever which moves instead.
That is an unnecessarily rude and mean comment.
You're the one calling people horses. I'm the one saying, "No, you're probably better than that." Which one of us is being rude here?
You. Clearly.
Neigh. It is you.
> Neigh.
HAHAHAHAHA in a discussion full of doomers, I really appreciate the guffaw that gave me. Thank you.
The job of a software engineer has always been to turn a real-world idea or problem into a precise specification a computer can execute. That specification used to be assembler, then higher-level languages like C and Python. Now it might be detailed Markdown files that AI turns into code.
So our understanding moves from the C/Python level to the Markdown specification level. We still need to understand exactly what we're building and why.
The hard part was never translating the spec into machine code. The hard part is defining the right spec. That is still the job of the software engineer.
I think we are very much in the below scenario:
"I'm a carpenter and electric power tools were just invented.
In the short term, people are asking if we need all these carpenters.
In the long term, people will want bigger houses and therefore the demand for carpenters will go up"
We are in the "short term" part of the above.
My real question is how long we spend in that period before moving on to the next one.
It is more like, "I'm a carpenter, and electric powertools that can also operate themselves, and run the sales department, and control a robot to change a lightbulb, and manage the entire company, and surround it with armed guards it also controls, just came out. I'm screwed."
The demand for carpentry goes up, not the demand for carpenters.
https://ourworldindata.org/grapher/agricultural-output-dolla...
https://ourworldindata.org/grapher/number-of-people-employed...
This is the Jevons Paradox.
Correct.
I'm just not sure exactly where we are on the "Jevons Paradox Adoption Curve"
I just wanted to point something out. I hear people say the AI revolution is the biggest one since the industrial revolution. I think it's the biggest revolution since the one that just preceded us, the information (computer) age. So we're still going fairly in order revolution wise.
The biggest revolutions since the Industrial Revolution were probably trains first and everything related to electricity later. Telegraph, radio, lightning, TVs spanning over about one century. And advances in health care, agriculture. But it's hard to tell without having lived through all those changes.
Then computers of course, but not as impactful as the other ones. And cars and mobile phones (mobile dumbed down computers) have been more important than AI so far IMHO.
I am already seeing more companies hire people who are developer-adjacent, especially now with AI-assistance.
I don't think software developers are going away, but likely headed more in-house, as each company can cheaply have bespoke software made and maintained.
I think the big problem is that nowadays, you can't simply switch professions any more, A.I. can do just about anything white collar, and there's only limited amount of blue collar jobs to go around.
So the future is going to be interesting.
Having managed programmers for more than 30 years, there are a LOT of programmers who just want to code. There are also a lot of programmers who went into programming because "it was a good, well paying job".
Doesn't really take much away from the article but a significant percentage of the developers who are unemployed are not losing their passion, they are worried about what job they will get to pay the bills.
I think the analogy of moving from horses to engines breaks down when going from code to AI as AI can quite plausibly work on itself. It's a fundamentally different game we're playing this time. And there were as many engine jobs as horse jobs (maybe more). The whole point of AI is super human productivity. Even the idea of just using AI means you need less humans because one human with AI can out-produce multiple humans without. It's very difficult to imagine a future where AI can do every job and things don't turn into a sci-fi dystopia.
My grandad on one side of the family was a market gardener. He didn't become a farmer.
My great grandad on the other side was a lighterman[0]. He didn't become a ship captain.
Sometimes professions really are just gone. In both cases in my family it was due to consolidation of resources and concentration of capital/power. Technology was just a catalyst for that.
[0] https://en.wikipedia.org/wiki/Lighterman
> Be willing to let go of the how to hold on to the why.
Without the how, I have no interest? The 'why' isn't the end result. I don't read a book so that I can say I've read the book, the purpose is the reading, not the state of 'having been read'.
> But it makes me wonder: what if what we loved about this job was never only the code?
This exactly. Look, I "love" code but only because of what it allows me to do and there is plenty of code/coding that I don't love. Not to be too hippy-dippy but I'm a builder, I like to build things. Build things for myself, build things for others. The code was _always_ a means to an end.
Sure, it was cool I had this skill that many people didn't have but I also used to have skill in managing the windows machines and things like the registry. When I first switched to macOS I felt lost, I had this "skill" that was no longer needed but after a while I realized that was a good thing, I wasn't wasting my time fiddling with crap that was just getting in the way of _using_ my computer.
Similarly I may mourn the loss of writing code but at the end of the day code was just a way for me to build, if there are better tools to build (LLMs) then I'm all for it.
I too love building things, but by building things I mean the act of actually building it. Thinking of a problem, devising a solution and then executing the steps to solve the problem, if that doesn’t work, debugging and trying another method and so on. I am not and will never be a fan of buying things, where I just ask for something and it gets made which is what you’re a fan of if you’re honest to yourself.
> Many of us, deep down, are driven more by the joy of making things, solving problems, and seeing people use what we built than by the act of coding itself. After all, many of us decided we wanted to be developers before learning to code, just to create video games or websites. The desire came before the code.
This is a big disconnect that I think has been at the root of many different reactions to LLMs lately. Here on HN I see a lot of "I hate what LLMs did to programming" and equally a lot of "LLMs are awesome, I've never had a better time". I think the disconnect is between people who build software for the end result (having users, the joy of completing a thing, etc) versus building it for the process itself (architecting, coding, debugging).
I'm mainly the kind who enjoys the process. I think of an awesome startup job I had that solved real problems for real customers including some really big firms, but I can't honestly say I ever cared about those problems or the customers. I loved the job because it was technically challenging with a lot of freedom, and the customers were great because they were a source of new interesting details to solve, as well as confirmations that the previous solutions worked.
As such LLMs make me sad because they effectively replace a lot of what I like doing. Doing a refactor to support a new feature, digging through the code to try and understand a bug. One of my strengths as a developer, I think, is that I am fast at familiarizing myself with unfamiliar codebases. That's also a somewhat redundant skill as Claude can answer "how does feature X work in this code" in a small fraction of the time it'd take me.
A few years ago (before the LLM craze) I moved to game dev as my main job, which makes a lot of this easier because I actually care about the end product and the customers (players), so that's a whole new layer of satisfaction in addition to the usual technical one.
Which isn't to say that it's all bad. I've used Claude to get a few things done where I mainly care about the final result and not the process. There was one such project I wanted to get done, I'd had the idea and the data for five or six years, but I found that the implementation would be time consuming, often manual work and much of it involving the kind of tech I don't really enjoy. So I never worked on that but last month, I did the whole thing in a few days with Claude and got exactly the result I wanted.
I've been seeing things the exact same way. There are those that want the end product as fast as possible and there are those that love the process and the struggle of getting there. The former group is the ones telling the latter group that they'll be left behind.
[dead]
[dead]
This reminds me of a TIME article by former Governor of Indiana, Eric Holcomb
https://time.com/7346751/ai-impact-american-manufacturing/
Curiosity helps a lot…. And the willingness to try out new things.
The difference here is that the car was not a centralized cloud service, owned by select few billionaires.
So the post does need an asterisk, referring to how it all hinges on availability of open weights and inference hardware.
Looks like the page got hugged to death...
https://archive.ph/51UP9
> The desire came before the code
Well put. You envision something you want, and then go through whatever it takes to make that a reality.
On the one hand, it makes sense to think of this as just another revolution as it keeps things familiar, keeps the ground underneath us intact and lets hope it goes this way.
On the other hand, if long horizon independent problem solving with almost zero human intervention or support is realized in these models 1-5 years down, then it will lead not to new tech, but essentially a new species of workers that can replace humans at all tasks, In that case, humans wouldn't really have the upper hand as an economical resource to get any work done.
Why is noone talking about the elephant in the room : at current LLM efficiency per W it's still around 1000x the one of our brain; interpolate and to replace us all we need to harness more than what the sun delivers...
Not a single comment about efficiency or consumption, only marketing stuff and bad analogies, not what it used to be HN... Where are all the nerds ?
> current
That's the problem: if everything stayed at the current level of development, we would be fine. There is no reason to think that will be the case. Intelligence per watt has been rapidly improving [0]
> 1000x
Coincidentally, that is exactly the efficiency gain promised by ASICs like Taalas [1]
[0] https://hazyresearch.stanford.edu/blog/2025-11-11-ipw
[1] https://taalas.com/
Funny thing for me, it was NEVER the code...
> Because, yes, we love to say AI is the biggest revolution since the Industrial Revolution, and that we’re going through a transformation unlike anything we’ve seen before. But that overlooks the farmers who watched the tractor roll in, or the farrier who watched the car arrive.
How does that overlook the farmers? They're alluded to right there in the prior sentence. I don't get this.
The mechanic analogy is rather useful. Today a mechanic is in a more squeezed space than we are, having to learn EVs and electronics, or accept to go out of business within the next 10 years.
Having worked with a few combustion engines I have no clue about how to repair an EV.
A fine example of how new technology is not better, since it closes the door on too many people.
Not only that cars are mode more and more impossible to repair by any random person by manufacturers on purpose.
sorry I think this is a false equivalency. This isn't the career path of programming getting replaced by a new technology. This is technology replacing career paths.
We're all familiar with the term "knowledge worker." What happens when that field is broadly removed? It will be a reverse-trend of telling all ex-coal miners to become software engineers. Except here's the rub: going from a software engineer to something requiring a human only for their dexterity is typically a massive earnings decrease. So, in effect, as Brian Merchant and others put it, this is not just a shift but a massive plan of wage theft.
Looking at it from the optimistic side as my bosses frame it: we're all moving up in the ranks! Well, my team already has a PM. She's already trying to vibe code her PRDs. Thankfully, right now, they suck. But they won't forever because all of us engineers are being forced to train our replacement. There can only be so many strategists, project managers, designers, etc. This isn't simply the technology changing and us adapting, this is a massive reduction in workforce requirements ACROSS industries. Hype bros thinking we'll all have jobs once the AI is good enough to replace us are simply hallucinating as bad as the frontier models of 2024.
There’s no demand for human dexterity anyway, and it’s not about money. I like doing intellectual work.
They are spending 1T to ruin the quality of work we do, also probably ruining our chances of securing a livelihood and then they want us to for some reason welcome it with open arms and openly say look how great this is and marvel at human obsolescence. Yeah no thanks.
Wowee, a post from the CTO of some AI slop company talking about how everyone getting replaced by the shiny LLM gods is a good thing, actually, how original!
How many of these do we really need? Nothing is happening, and a lot of people are upset about it.
They wanted to make money, or at least be a part of history during a transformational era. No dice.
There are two key elements in the current AI fearmongering:
1) People think they can predict the future
2) People think humans can't adapt
The number of disruptive "game changing" technologies throughout history is immense. There's no question some people get negatively affected, but humanity as a whole has prospered because of these disruptions.
Literally every time, including during the industrial revolution, people thought it was game over for society, good manners, the youth, the planet etc.
Keep cool and try to adapt to the inevitable change. The next generation will certainly do it and leave you behind if you stay in a corner denying the new reality.
Adapt, sure, but is my life better or worse after adapting? It's a fair question and a fair worry. The goal of most AI companies seems to be to replace some amount of labor with AI. Look how that worked out for our factory workers. They adapted by getting jobs at Walmart and Dollar General, or scraping by in the gig economy.
> They adapted by getting jobs at Walmart and Dollar General, or scraping by in the gig economy.
The tech crowd used to say that only the "dumb" ones did that. They said those who were "smart like us" were much more adaptable, learning to code and whatnot to find new, often even better careers. It will be interesting to see if the emperor has clothes.
Can you not see how you're doing the same thing now? Not all people in tech so dismissed the struggles of others -- but you are.
No, and obviously neither can you, else you would have explained yourself. Best to not post while under the influence of silly emotions. That is what those tech people who called those manufacturing people "dumb" did and, like this, wasn't useful.
Silly emotions? I'm not the one gleefully celebrating the ruin of others (including families). You've chosen to 'other' a broad group of people with diverse experiences and backgrounds. "Those tech people" certainly include some people who called manufacturers 'dumb', but it also includes many (most?) who didn't, and even some who used to be "manufacturing people".
You say "who cares if they suffer? They're obviously all just bad people". The people who attacked blue collar workers said "who cares if they suffer? They're obviously all just dumb people (and thus bad)." So the bulk of people who you decry are not guilty of what you claim. But you are!
> You say "who cares if they suffer? They're obviously all just bad people".
This comment aside, I do not see that text anywhere, lest my own comments. Get back to us when you return to reality.
> The people who attacked blue collar workers said "who cares if they suffer? They're obviously all just dumb people (and thus bad)."
Whether or not calling them "dumb" was justified, there no implication of "who cares if they suffer". In fact, in the context I remember, those comments were made around trying to figure out how to help those who were deemed "dumb".
Why is it whenever people try to downplay the efficacy of LLM’s evangelists go “there’s literally never been anything like this,” but the moment people critique them or try to weigh their impact on society the response is “well it’s nothing we haven’t seen before”?
The amount of money flowing and the sheer scale of all of this stuff is without precedent. We are seeing governments from the local to federal level literally shift their economies because of all the speculation. We are clearly experiencing a moment of sorts, and anybody who is worried about what the fallout will be is just handwaved away as ignorant fear mongers. Unless you are the head of these companies doing it in a disingenuous way to pump up your valuation, then suddenly it’s legitimate - but not so legitimate that we should do anything besides let the US government build a moat for them!
> The amount of money flowing and the sheer scale of all of this stuff is without precedent. We are seeing governments from the local to federal level literally shift their economies because of all the speculation.
Yes, the absolute numbers might now read TRILLION instead of BILLION or MILLION, but the concept of "Technology so impactful governments shift their economies" is in no way new, and has already happened a bunch of times before in humanity's timeline.
I guess that's why you're seeing some of us less stressed out about it, as again, scale is different, because scale today is different in everything, but relatively, we're just repeating another of those "Technology changes everything" milestones that we've done before.
We have never seen governments essentially leverage their economies so heavily, so quickly, for a speculative technology. It is literally propping up a significant portion of the US economy right now.
When have we seen anything like this?
> When have we seen anything like this?
The US canal and railroad boom in the 1800s definitely had exactly the same thing happening, just compared to our scale today where everything is "more" and "bigger", it looks comparatively small. But in their world, back then, it was basically the same thing, just with canals and railroads. Indiana almost financially ruined itself because of their "infrastructure program". The government got involved heavily with the "1862 Pacific Railway Act" as well.
And yes, easy to say now that "Duh, that's essentially infrastructure" but at the time, that wasn't as obvious to everyone, as and railroad were newfangled technology, especially deployed like that, on a scale never seen before.
> definitely had exactly the same thing happening, just compared to our scale today where everything is "more" and "bigger",
And faster. More, bigger, faster. Which substantively changes the situation the same way a musket stops venn diagraming as a “firearm” at some point when compared to a 50-round automatic rifle.
You're mixing a bunch of different "people" in the same bag. I'm not the one saying "there's never been anything like this before".
The investment is massive, but not bigger than other projects by an order of magnitude. Major wars were bigger than AI. The dot com bubble was 2-3T dollars.
If we’re comparing war between entire nations to speculative economic investment than I’m not sure I need to make any more points.
There may not be so much in the way of soldiers in the field guns blazing, but we are very much seeing a 'cold war' emerging with AI at the centre of it. Call it speculative, but the incentives for nations to spend are essentially the same in cold and hot wars.
Because they're liars. Plain and simple. Lying is a virtue as long as the lie is powerful enough to sustain attention.
I wish the author had gone deeper
What a refreshingly positive read
That is inherent to propaganda.
No it isn’t, fear mongering is way more common for propaganda. What makes you think it’s the case here?
> Many of us, deep down, are driven more by the joy of making things, solving problems, and seeing people use what we built than by the act of coding itself.
I LOVE playing "product designer" with LLM. The thing I normally hate and never do, write and update "design docs" being my only lever is awesome. I do already hate how it seems to turn development into a much less social activity. It's great that we don't need other people and a lot of time for some things now, but it's also not.
But more importantly, because I also like solving problems and seeing people thrive, I can't overlook that it's all based on stuff taken without consent, how artists were made "obsolete" by what plasters me with slop every day (already, and we didn't even get started). And they're not even going "okay, now that we proved it can work with stolen stuff, let's do it proper". No, this is "the new world now", just keep pushing and squeezing and extract more "value". So wtf? Well, I like solving problems, I love games! One cool game is "catch the thief", another is "call out the sociopath". So let's?
But we can't, because we're not building this after deliberation and discussion, we're being herded into it and then make rationalizations for foregone conclusions made in our absence against our interests. Building systems nobody understands that keep making more systems nobody even tries to understand is not some hyper-efficient future, it's no future at all, and nobody with power (however wishy-washy that phrase is) wants to live under any of that, you can be assured.
So this all seems more like an airlock to me, an abattoir, not a spaceship that just happens to be kinda dirty and stinky because the focus is on so much grander things. How insincere and stupid so much of it is just can't be overlooked, and very much unlike other technological advancements where the results speak for themselves, where the results (or the improved process to get them), pull something, make it inevitably seep everywhere it is useful. There is so much pushing and hyping, and so precious little good stuff to make for the really bad stuff, all the slop that is produced and clogs up inter-human channels.
This is not even progress, this is just subversion, that's my thesis. It's deliberately driven for the easily predictable bad effects it will have.
Humans ourselves are a technology. We have incredibly flexible brains and dexterous hands. Up to now, we could always do things that robots couldn't. That was where our job security came from.
But what happens when a robot can do anything a human can? When a cheap machine can do anything you can do but faster and better, what makes you different from a horse in the 1920s?
This is what the article misses and what people have trouble accepting. Yea, this is a technological revolution, but unlike the Industrial Revolution, you can't just upskill your way out of it.
> But what happens when a robot can do anything a human can? When a cheap machine can do anything you can do but faster and better, what makes you different from a horse in the 1920s?
I've already started to feel this.
One thing that made us special was our intelligence. We were, and were always going to be (unless some alien form of life arrived), the most intelligent things on earth. Is that still true today?
We should start with the assumption that humans are valuable and deserve to live with dignity. Any philosophy that doesn’t have that at its center is broken.
We should, but to date anyone that wanted to be treated as though they were valuable and deserved dignity needed some way to apply pressure to ensure that.
I'm not optimistic that this will change in the future -- but, I am watching automation erode the amount of pressure that humans can apply.
Yes but same should have been true of animals and humans as a whole did not afford dignity or value to animals. Maybe we will get a taste of our own medicine this time.
I'm curious where you are drawing the line. Why is a human different from a ape/dog/chicken now? I wonder: do they not deserve to live with dignity?
Depends on who you ask. Some people might put it at opposable thumbs. Others might ask how many legs one has. Or language, or ability to count (crows can count up to 30). Memory (elephants have memories that go back over a decade).
Perhaps. But are humans going to get what they deserve?
I have some bad news for you about how our society actually functions...
Very often the justification for why these things aren't terribly disruptive is that the very best of a certain trade or field will still have a useful skillset. Like the best graphic designers in the world are still more creative and skilled than AI, and the best writers are still better, the best programmers are still better, just that they're all more expensive and slower.
But we're not taking into account how it affects the mediocre or average people in these fields. Roughly half of all of these people are below average. Maybe if you're the best you'll always have a job, but what about the guy who legitimately struggled to finish high school? The person who genuinely doesn't have the capacity to earn much more than minimum wage?
We're currently at the part of the story where chess AI can beat amateurs and is rapidly improving. Kasparov's not worried, he's an expert. He's more creative and skilled than an AI.
Why would these machines stop improving at average this time?
I didn't say they weren't improving. My point is that if our justification is comparing the top 10% of performers to AI to justify human performance, that speaks poorly for the replacement of the bottom 10% of humans.
And why not the top 1% or the top 0.1% or the top 0.001%? Either AI stops improving, or we're in for a grim future for people.
I weep for people whose scientistic reductionism has led to such devaluation of humanity. Robots don’t do anything and don’t think anything. Humans do things with and through them. What makes us special is that we exist and love one another.
> Humans ourselves are a technology.
Grim.
https://www.vatican.va/content/leo-xiv/en/encyclicals/docume...
The economy long moved past the things we can do, hence "bullshit jobs". Our job security has come from our willingness to "look the other way", allowing the job creators to live out lives the rest of us can only imagine. If they didn't placate our insatiable desire to work, that wouldn't last long.
[dead]
The western world, the US in particular, already has outrageous abundance.
And yet problems of despair, poverty and marginalization continue to exist.
5-10% of people experience some level of food insecurity. Is that really because of a lack of resources??
It seems to me like we have a resource allocation problem, not a resource generation problem.
This is not an argument for some form a socialism. But instead an argument that progress in the field of social organization would be more productive than progress in the field of technological disruption.
And I love tech.
"My family" - a nice anecdotal frame. No worries here. After all, labor savers for production has evolved for millennia. And you should only worry about the impact to yourself, and be glad your lot is better than your ancestors'.
But our society today has very extensive systems with very few deciders, who are largely enmeshed in social and legal constraints to ensure our reliance is safe and effective (at least in the large/long run).
The concern is about that constraint system being displaced - minimally with decider over-reliance on AI (still subject to guardrails?), or maximally with AI itself (no internal guardrails). Slop (in software, medical services, and military targeting) is just the symptom.
That replacement/shift has already been a corrosive side-effect of capitalism, as elites slowly cede loyalty and professional standards to competitive pressure, but even now most constraints allow replacement of bad actors (if demonstrable). AI reliance just makes that shift irreversible. People no longer have incentives or opportunities to spend the decades required to lead effectively, so organizations could never revert the heartless efficient machine elites back to the available incompetent and oblivious humans.
Maybe the result is just wealth transfer, or health, but I'd expect the high-value use to be in a military where very few people can combat many (as is happening in Ukraine now), making the military a preferred option for anyone with resources.
The distribution of (and checks on) responsibility - up, down, across companies and political systems - was intended to constrain agents to organizational goals, but it effectively kills any internal change by requiring near-unanimity and overcoming people's dependence on their jobs. So even if a super-majority of people agreed on some negative projection of AI usage AND had the resources to put work and investments at risk, it still wouldn't be enough to change this trajectory.
More AI deepthroating with the usual psychological-care veneer.
No, the difference this time is that the goal is to replace all coding, and perhaps all office work. Two obvious problems:
1. All of Office Labor has nowhere to go to—how to translate thirty years of experience staring at a screen into something else? And will there even be other jobs for them?
2. No labor leverage since there is no labor in that sector and the capitalists just capture all of the value for free (with an initial investment in CAPEX...)
How does this compare to horses and cars? You still have to get around.
Granted these are the maximalist AI Cretin goals and I’m also influenced by their relentless propaganda campaign.
> Many of us, deep down, are driven more by the joy of making things, solving problems, and seeing people use what we built than by the act of coding itself. After all, many of us decided we wanted to be developers before learning to code, just to create video games or websites. The desire came before the code.
News to me. As long as I was in this wider culture, coders were sneering at people who were in it “for the money”[1] and not for the intrinsic pleasure of very specifically typing out and testing code. And at ideas people. But as soon as these programs hit the mainstream, oh the most vocal ones just never cared about coding itself and they were in fact passionate about all of those business value-adds they were adding.
And managers? Clueless people herders. But what are clankers comparing GenAI coding to? Management. And this is enjoyable to you?
People will yell Goomba Fallacy. Of course maybe that whole old guard was just displaced.
[1] Working in order to survive: what a concept.
in about 15-20 years when all office jobs are gone and we're only talking to machines there is going to be "ai alignment and safety specialists" doing the same jobs in smaller offices for less pay than customer support does today. they'll be fixing "misaligned" models and penalizing citizens for deviating (see credit scores, penal system today) all from the comfort of their home.
they'll be more afraid of terrorism, black people and muslims than ever. and lets not get started about socialists!
however you spin this it is going to deteriorate your average job opportunity at least in the us. the us will be where China was 50 years ago: worse workers rights, worse political participation, more low income jobs working for "miltech superintelligence corp".
what hegseth does to your military, amazon-ifying it is what is also happening within your political system stripping you of your civil rights.
at least china had an economical goal furthering the nation state. maga has... what? sucking up to elons, zucks, bezos and other super rich weirdos in the hope of resurrecting the big american empire again. it's devastatingly stupid, really.
[flagged]
[dead]
[dead]
[flagged]
[dead]
[dead]
[dead]
Aside the fact the author sells ai services, training, whatever i can not understand the view how knowledge workers will be replaced while every employee is now forced and expected to work 10-fold and hiring is halted because shortsighted deciders and ... money.
The effect on the market is pure GREED and wishful thinking as none of the people in charge of HR or decision making have any idea how this technology actually works.
Also - why is everyone comparing this to the industrial revolution? And not f.e. invention of vaccines. No it HAS to be the big big story or you cannot make money with it.
All while the actual results are well, great. Helpful, sure. But so are vaccines!
Don't buy into this bs, any service or course. d2l.ai is free and wont be stupidly useless next month like whatever hype we sell this week.
As the doom stories lose novelty now we get the feel good "my grandpa" stories.
Thank you for your comment. I didn't intend to tell a feel good story. I'm just a seed startup founder, I'm not DHH! I'm just sharing a personal story that helps me navigating into the storm. There is no message here, I guess I hadn't been clear enough.
People are flagging in real time. Maybe the grandfather would have thought you are cowards and frauds who abuse him to push a vibe coding company.
"what if what we loved about this job was never only the code?"
The phrasing is very LLM. Apparently what you loved about writing blog posts wasn't the writing either, since looks like you are having AI do it too!