So, I really felt like more people should be reading Nassim Taleb's Incerto series of books. A lot of the issues that fall out of AI he dealt with in his books like ten years ago.
He gives one the best pieces of advice I've ever heard: if you are going to do something for a living, make sure it is NOT scalable.
If you do something that isn't really scalable, like being a welder or a tailor, then you only have to compete against the tailors in your neighborhood, and you can easily find a neighborhood that doesn't have a tailor. If you're building a scalable product, you'll always be competing against the best, most well funded, smartest people in the room.
Everyone here has grown up in the birth of the internet -- a once in ever event -- where building something scalable was just there for the taking. That's never going to exist again basically.
That seems like reasonable advice until you realize you have no idea what will and will not scale in a few years, and there's only so many tailors/plumbers/welders/etc.
What it really comes down to is these options as more work gets automated:
1) New jobs doing different things that weren't done before
2) Same jobs but shorter hours so "full time" with a salary to match starts to look more like 4 days, or 5 hours/day, or something.
3) Lots of unemployment
These can happen in a lot of different combinations, they can come wrapped up in different ways, and unequally for different segments of the workforce, but there's limited elasticity in most areas where additional people piling into the field would create more demand rather than glut the supply to the supplier's detriment.
I keep hearing about the potential of "new jobs" coming from ai but can anyone actually describe one? My gut says they will be something similar to converting middle class knowledge workers to do DoorDash drivers or trained artists becoming dog groomers. What a cool future, at least my parent's can watch racist ai slop videos on their iPads.
It's simply a case of looking back and deciding this technical revolution is identical to the ones preceding it. Thus jobs destroyed will be replaced with new jobs, it always happens that way.
Of course past performance is no guarantee of future success...
95% of human farmers lost their jobs because of industrial revolution. What happened then? No jobs were created and we still have 95% unemployment, right?
This is assuming the conclusion. The entire question is whether we are the horses or every other example of humans in the past who found other employment that was inconceivable previous to the technological revolution that rendered their old job irrelevant.
The point of doing something non-scalable is that you can enter and exit the market fairly easily. You don't need to be a tailor your whole life. You can make a living as a barber, electrician, teacher, or nurse.
I'm not saying it's easy! It's hard as hell. It sucks when your job gets automated. I'm just saying that aiming for something non-scalable means you're not always tilting at windmills, and the game can't be rigged against you.
Switching jobs from electrician to teacher to nurse will take around 3 to 5 years of education or apprenticeship in most countries. It also requires new licenses or certificates if you ever move country.
I'm trying to respond to this stuff in good faith. Yes, I agree with you. I don't see how this is relevant to the argument. If you are in a non-scalable industry that gets taken over by technology, that sucks. My point was being in the startup game in the first place.
"Becoming a teacher" takes years.
"Becoming a successful scalable business" has no known time frame. It either happens or it doesn't. And whether it does is not particularly correlated to how much time or effort you spend on it.
Almost typed the same thing in different words at the same time :)
Your above point can not be upvoted enough.
You do have to wonder what some people are thinking ;)
Good message from Taleb too, he may not be the most likeable or agreeable person in the world, but he is quite brilliant at synthesizing logical meaning from the inputs he does have.
People not liking Taleb and then dismissing him because he can be rude has always been one of the strangest things that so many people do.
I remember about halfway through Skin in the Game, after being put off by most of what I'd been reading, that the gestalt shift happened and it finally clicked. It's probably one of the best texts I've ever read on a specific type of conservatism. I don't like that I agreed with it, but the guy makes a very, very convincing argument.
All of those professions you've listed require about half a decade of dedicated training to be legally allowed to practice. For example an electrician takes like 7 years to become qualified, that's a full time apprenticeship, and it pays badly in the meantime.
A fact endlessly annoying to electrical engineers who legally can design their houses power system but not work on it.
(I mean I think a barber is quicker, but one of that list is also not like the others)
There are a lot of code issues you can run into when actually installing and repairing wiring that your average EE wouldn’t know much about at all. And just because you can design the power system for a building doesn’t mean you have any how to fish romex through a wall.
That’s like saying a mechanical engineer should necessarily be able to work as a machinist. Some of them can for sure, but it’s not something they are required to learn.
Also in many places anyone can do electrical work on their own house.
> For example an electrician takes like 7 years to become qualified
Using a similar playbook for what happened to SE (loads of good ~free info, bootcampification) you can drastically reduce the time to be economically useful, but that can only go for so long due to saturation. Especially residential electric work (or other trades) is no witchcraft that would require anywhere near several years of education/practice. And yeah, the legal system would have to cooperate, which will only happen when the pressure becomes high enough (see e.g. debates around "Meisterpflicht" in Germany).
That said if AI continues to progress like it currently does we may not get to that part.
I'm not trying to say 'everything is fine, nothing bad is happening a world of recurring technology and industrial revolutions.' It's not. The way things are set up is bad.
My point is the author writes a column about how GPTs are ruining the ability for people to make scalable products, because when everyone can make one, nobody cares... my point is that that's not the result of GPTs. It's a result of survivorship bias skewing how we look at things.
When your business is a flywheel than needs to be running to provide a benefit to each user, then getting that flywheel running is a huge problem. The vast majority of non-scalable businesses, almost by definition, provide each customer with a benefit regardless of whether anyone else uses it. That is how you create basic, word-of-mouth, free "earned" marketing.
Being a tailor is scalable, that's way there are way, way more cheap machine produced clothes today than in the past. Surely he did not miss that the textile industry was at the core of industrial revolution. So being a tailor is more like a post-scaling job - the automation has already happened and now there are only remnants left.
But how can you be sure a job is peak-automated? A few years ago, I would have said musicians are post-scaling - way fewer musicians jobs now that you can play recorded music. But it looks like generative AI will hit musicians again. Can some of welding be automated? Probably.
Welding is being heavily targeted for automation, apart from pressure vessels etc, most welding can be automated now a days , very soon ( months? ) every welding can be automated.
You're thinking of factory welding, manufacturing or maybe repetitive pipeline welding and thinks like that.
It'll be a while until a robot disassembles a trailer enough to remove the bent axle, cleans off all the paint and rust, bends the trailer back into shape where needed, cuts a custom support to makeup for some lost strength, welds it in, primers it, paints it, and assembles the trailer.
If automation excels sufficiently, robot will replace bent axle with new one automatically, repair cost will be usually higher than replace cost if all things are automated
I mean a tailor who adjusts clothes and occasionally makes something bespoke.
Tailors typically operate a launderette and act as middlemen to a local dry cleaner.
I’m not talking about a fancy man making clothes for rich people, I’m talking about the talented old lady in you neighborhood who adjusts your clothing for $50 and runs a wash and fold.
How many people visit either kind of business these days? I'm almost 35 and have never once gotten any garment tailored. I think one reason is that clothes are so cheap you can just keep looking and you'll find something that fits, which seems to be what most people do nowadays.
Try it. You might be missing out! One of the benefits of getting all your trousers from a tailor made to your size is you never have to waste time trying on trousers, or waste time ordering-returning items.
Yes, but this is simply the remnants of the old tailor occupation, post automation. The talented old lady would have had a lot more business in clothes making in the past, no need for a wash & fold.
My mom altered clothing when I was younger - and she darned socks too. My M-I-L still sews the occasional seam for pants that are too long and were cut for my wife or daughters.
But me? I buy a pair of $30 jeans at Costco. If they don't fit great, I buy a different pair of $30 jeans. I don't spend $50 to have them altered, or take it to a laundrette. If it can't be washed in our home washer/dryer, I don't buy it. And these days, when a sock gets a hole? I throw it away.
This is an effective strategy if you're a fit model, or close to one. If you're within the standard deviation of the sizing chart, you'll probably do fine most of the time.
I like golf. Most people use a standard shaft. In fact, that shaft length is standard because most people use it. That doesn't mean there isn't an entire industry for "golf fittings" because "most" people isn't even close to everyone.
Well, that's why there are so few people doing this full-time any more.
But at least they're not going to disappear completely.
And now there's nowhere to go but up :)
They have had the internet a while too, keep in mind a select few have gone viral on Etsy while so many SaaS things don't return a fraction of their potential.
And plenty of alterations going on all the time after all the automation dust had settled manufacturing most fashions, a lot less manual work is of course being done but it's still everywhere. You do have to be good or you're not going to do half as well as you could though.
The thing is, automation should be expected to slow or stall sooner or later, automation's not suitable for every little bit of welding or sewing that needs to keep going on. Only the most suitable, of course ;)
These are just random examples, if you want to make absolutely sure you won't be automated away by the internet, build a valuable skill that doesn't depend on the internet at all, nor look anywhere near the places where automation is emerging that it wasn't doing before.
If you eventually figure out how to automate that skill it would be something.
Just like the internet though, there can be extra credit for being first :)
One of the most valuable things to be able to build single-handedly is something that can not be mass-produced by any stretch of the imagination.
You might stick with that alone, or pivot to something with more of a financial upside, but you would always have something to fall back on if needed. Plus give you less worry about taking financial risks than you would have been, considering the same resources and/or capital to work with.
And on a regular basis revisit how far you can stretch your imagination to see if your baseline fallback still doesn't look like it will ever be automated in a way that would effect you.
I used to look at my local dry cleaner and think that it had to be the most stable business going - it couldn’t be replaced by a computer. Then Covid hit, work from home took off, and the dry cleaner went out of business. Computers came for the dry cleaner, not by cleaning clothes but by eliminating the need for dress codes.
> if you are going to do something for a living, make sure it is NOT scalable
You need to consider both horizontal and vertical scaling. Being a bespoke tailor may not scale vertically, but it can scale horizontally. If you have too many people pick up tailoring, you might run out of neighborhoods without competitors.
I wouldn't say it's not really a bad thing, I think it's a very good thing. There are many people now making incredibly niche products that have very good lives - making more than enough money for themselves doing interesting work engaging with customers that are passionate about their field.
Given the success of copyright cultures in fending off the most blatant, unambiguous case of copyright violation in human history, (AI), I would not put your money there.
One early approach when starting to build a reputation is to work as a subcontractor to a less-big company who has already gone through everything to be on the approved-vendor list of the massive corporation.
Even better if the less-big company is a private company and the biggie is public.
A private company will be able to figure out the benefit from your work better than most, and hopefully profit from it.
I like it when every time I invoice a client, they are making money at the same time, whether they are making it from me is not always necessary if I am participating in an overall money-making process they are going through. Especially routinely. It's just fine to be a very small participant in lots of activity going on between the big-shots.
Big companies can have so much work and be able to pay so well that your position can be to merely absorb the overflow from the primary contractor, even if you do not yet have a unique offering.
Then it's good to build into your exclusive-but-related niche, which would be intended for a different division of the biggie and expect it to languish unless your already-established client can use that too.
Eventually after you finally get to meet the right person in the target corporate division, it will be after you have already been doing work for that same corporation through the approved primary contractor.
By this point you've already been doing critical work for that corporation exactly like they need, that you can be proud of and point to, and it can be a whole lot more likely to become an approved vendor yourself. From which position you can finally negotiate fees directly with the most well-heeled corporate source, this would be the first time if you were only working for other primary contractors until then.
Ideally you will then be invoicing a different office of the same corporation that you were subcontracting under, there will be no conflict of interest, and you can continue working for the original primary contractor too.
And your most promising niche finally gets to launch with about as much upside as it can get.
Welding is coming along. Ever heard of Path Robotics? That's high-mix, high-complexity welds. (There are a lot of other, less-sophisticated welding robots out there.) The biggest moat for a welder right now is special certifications to be able to do welds on submarines, etc.
I think that delivery services give you bigger market, but not intrinsic scale. You're still limited by kitchen size, staff numbers, and raw hours you can put into the food.
You can scale the system (say Subway, or even smaller chains like Burger Fuel), but also reasonably choose _not_ to scale and still do incredibly well (like Michelin star restaurants, or the myriad of hyper-famous-locally Japanese eateries, or Fergburger in Queenstown).
Someone scaling their own restaurant on the other side of the world won't necessarily out compete out (and in the overwhelming majority of cases with have no impact at all). Despite fast food joints all over the place, I still see small cafes, individual eateries, etc performing well (I mean, as well as hospitality can be).
Maybe it's worth expanding the definition of > make sure it is NOT scalable < to include 'make sure it's not automatable'?
The problem is that any market can be taken over by someone with deep pockets (or investors making a pile of money). They just make sure they are the "go-to place" for consumers to access the market. By marketing the hell out of it, by making apps, abusing power in other markets (see platforms), etc.
I'm excited for a future where the technologist is like the tailor in their community. Scaling software has created a host of 'product traps' and there is no need for that for the vast number of activities people do aided by technology.
There are much less expensive ways of playing the lottery than investing a year of you life into an unprofitable product that must be scaled to be successful, and only then hoping gets bought by a direct competitor.
If we get AGI and fully autonomous robot assistants, we'll live in a post scarcity world like Star Trek, or somebody in control the robots will use them to enslave all of humanity... so... high variance outcome could go either way.
IMO literally automating the job isn't the only possible scenario -- already in the weightlifting space coaches are supplying skills and system prompts to llms that do their training for them, massively raising the number of students they can train at at time -- at some point that turns training into a zero sum game
Even if it does, there will probably be a prolonged economic period where robots are doing dangerous/messy stuff like welding, plumbing but there is a human master guiding them from a few yards away, via prompts, controllers, etc. More of a semi-autonomous power tool than an fully autonomous master that is delivered by drone on-demand. Scalability is still a ways off.
A big point of seeing a tailor is getting yourself fitted for custom clothing that is specifically made just for you. As someone who's bought $200 off-the-rack suits and $2,000 tailor-made suits, there's a world of difference between the two, especially when you have an atypical body type.
(Granted, to the main point, I still think a tailor could be automated in some distant future, but we'll need robots to perform physical interactions, not just software.)
Tailors are a niche thing for weirdos, now. It's not exactly a growth market. Most folks only wear a suit to weddings and funerals, and maybe job interviews. They have basically no need for more than two suits, and many try to get by with just one (in black, probably). Lots don't own one at all, maybe just a cheap fused-construction blazer or two, if even that. Outright bespoke clothes are a niche of a niche.
Normal people wear clothes containing minimum 2% elastic and perhaps never, ever visit a tailor in their whole lives, except maybe one at a tux rental place or a wedding dress store, for their own wedding. If they repair clothes, it's sewing on the odd button at home or using iron-on denim patches. Past that, it's just not worth fixing, normal folks' clothes are so cheap.
The whole market for tailors is practically an affectation. It's not serving much actual need any more, not from the perspective of the overwhelming majority of people who are happy with stretch-denim jeans and polyester sportswear jackets and such. It's basically 99% of the way to being an obsolete job, kept from total death by a few enthusiasts. Only a bit more lively than the market for, say, authentic regency-era footwear or something like that.
Yes, I am one of the people who has a preferred tailor who can do more than just let trousers waists out. I also know where the nearest cobbler is. That’s not normal, though.
A dead industry often doesn’t entirely disappear, it just shrinks a bunch and comes to rely entirely on enthusiasts or very rare actual need, rather than broad need or appeal. Consider the draft horse breeder, or the carriage driver. There’s a market for both professions! But they’re itty-bitty. The day-to-day need for both is gone.
Tailoring is hovering right in the edge of that kind of status, today. It’s dying, killed by $10-30 shirts and $20-50 trousers and $50-100 jackets all from largely synthetic materials, and a society that no longer expects anyone to wear anything “fancier” outside certain events.
I mean, outside very unusual circles, dinner jackets are essentially ceremonial costume-wear, and business suits aren’t far behind on that track. You gonna wear a tailored wool hacking jacket or breathable linen Norfolk suit on your camping trip, or a bunch of polyester and nylon stuff from REI? LOL. All the situational tailored clothing but the business suit and blazer are near-extinct unless you want to look like a cosplayer, and those are on borrowed time.
Yes, your message is coming from the pov of economics and business, as makes sense in this thread! That's my mistake, I took your message more sentimentally. I've used tailoring probably 5 times in my life, with the only recurring need being to hem pants.
"There is no money in tailoring" seems right. It's the "not all things need to make maximum $$$" that I speak to. You didn't pick this fight though, I did heh.
My (successful) friend tells me all about how amazing it is to collect very expensive watches. I just need to be a "watch guy" and I'll come to understand. Once my eyes returned from rolling out of my head, I did concede a great point he made: there is no reason for watch makers to exist anymore. The fantastically amazing history and evolution of time-keeping and personal time-pieces is now purely supported by rich people that care to subsidize the art form. And so, maybe I really do aspire to be a watch guy after all... hmm.
A romantic perspective I still try and hold myself, however the point about the watch and the cloth and the dwindling appreciation for such is presently experienced in reference to decades or centuries of disruption and are intrinsically tied to the demand of attention. I don't trust the acceleration will leave much, but I am continuing to paint and taking writing more seriously in great fear of the time scales we are navigating today. I find myself confronted with nihilism in so many facets of my life but perhaps this is simply the smell of the air in my particular milieu.
A huge part of the tailoring business are making small adjustments to cheap clothing to get them 90% of the way to bespoke.
If you’ve never done it, I strongly recommend getting your jackets tailored. Even a casual jacket will fit and look non-trivially better for $50-$100 and an afternoon at your local tailor. You can even get things like cycling gear tailored.
Rich people still get suits custom made specifically to their measurements and preferences. They cost about $20,000 USD. It would be cool to have this process automated and affordable to the masses.
Absolutely. Design parametric families of patterns, 3d-scan the person, let customer adjust with live preview, laser cut, then fully automated or low-skill assembly. Probably not currently economical like many things involving physical world manipulation, but without obvious roadblocks.
> if you are going to do something for a living, make sure it is NOT scalable.
Great advice but difficult to action though.
I mean 10 years back I'd have thought programming is that thing which is not scalable. I had every reason to believe that. It required skill, experience, ability to stay current, grit for debugging hard stuff. Much of it can be automated now.
What can I pick now for a living that is not scalable today that some future technology would automate it just as easily.
> When creation was hard, skill was the differentiator: you had to actually be good to make something worth showing. Now the barrier is near zero, so you need reach. Reach costs money or it costs years. Probably both.
Creation has progressively been getting easier since the invention of the computer, it is not a new phenomena. This naturally pushes the boundary on what needs to be delivered in order to find paying customers. In other words, creation still is "hard" if you want to succeed.
> I launched something last week. 14 people signed up — no ads, just a couple of posts. 14 real people who didn't have to. That number is tiny and it felt like something. Then I sat down to think about what it would take to grow it and I couldn't look at that math for very long.
This applies to 90+% of founders that have ever launched something. The hard part comes from continuing to push forward when you experience this (which you will over and over). It sounds like the author expects that what was hard suddenly should be easy.
I didn't read the article, but yes, going from 0-1,1-10 is really hard and really rewarding. And it got easier with the Internet. Going from 10-1k and 1k-1M is a different ball-game. Always was.
The dream of running my own company got me to learn programming. 20 years later I'm an employee at a company still dreaming of running my own company. But now I can realize that reality and dreams are not the same and that's ok. As in, I probably really don't want to run my own company. We'll see =P
>> The dream of running my own company got me to learn programming. 20 years later I'm an employee at a company still dreaming of running my own company. But now I can realize that reality and dreams are not the same and that's ok. As in, I probably really don't want to run my own company. We'll see =P
To me having one own's company was just a means to the end: making enough money to live comfortably without the need to get a job ever again for the rest of my life. I too learned programming as a means to achieve that end but eventually realized that I don't need a company if I can short-circuit the path to money. By switching to the right domain - finance, where what I learn might be eventually put to use directly by investing capital into profitable trading strategies.
I work in this domain since almost 20 years and can tell you, noone's gonna risk a billion dollars on crap vibe coded by AI. I wrote before, I don't know what crack these AI people are smoking but when there's real stakes at play, they don't play around with toys. And AI in programming is a toy. The unlikely triumph of "Can I haz teh codez?" CTRL+C / CTRL+V "prompt experts" (mocking it, lol) strategy on Stack Overflow, along with the people who employ it.
I'm not worried about MY particular future in this industry. I'm not worried that AI is gonna replace me, us, or write anything significant here at all in the foreseeable future until it fucking evolves into AGI which is somewhere 5000 years from now, optimistically.
The party's gotta come to an end really soon along with the figures on how much money AI makes versus it's real utility - which is, simply stated, "a toy".
Something I've noticed a lot with Twitch streamers and YouTubers is that many of them want the outcome and are prepared to do the work, but want some sort of guarantee of success. It's very difficult to really sell to people that you will work very hard for a while and there's absolutely no promise that you'll ever have more than 14 subscribers. That's simply the core risk of entrepreneurship.
Also, as someone who's main source of income is a YouTube channel: there is a type of threshold effect, where your videos are not good enough to watch until one day they suddenly are.
This means that until you reach that threshold, it feels like you're not making progress, cause every video just gets the same result (no views). Even if below the surface, you're slowly inching closer to that moment where your videos will actually be watched.
I have the impression it's more of a "bait and switch" thing. People get into streaming thinking they can just play games and make money, but then they realize nobody is going to watch someone play game X, they all want to watch game Y, because game Y is popular right now among the people who have the most time to play games and watch streams (kids). So they enter the industry thinking they can just do whatever they want, and quickly realize they have to do infinite things they don't want to actually reach the level of popularity they need.
> Creation has progressively been getting easier since the invention of the computer, it is not a new phenomena. This naturally pushes the boundary on what needs to be delivered in order to find paying customers. In other words, creation still is "hard" if you want to succeed.
Only for developers. Outside of software creation is still hard. Global markets giving access to excellent manufacturing sure does help, but software is a bubble.
Creating marketing material has certainly gotten easier as well, it used to require a lot of work to create these spam pamphlets and company documents but today its trivial. Of course those are worthless to society so didn't help GDP but it filled our society with advertisements and spam and filled companies with worthless documents since now nobody thinks before making one.
The thing is, the barrier isn't near zero. The time to reach an MVP has just decreased. But you still very much need expertise, strategy, etc. to deliver something worthwhile. The bar has just increased.
Apple Pay (launched around 12 years ago, and likely costing billions) is exactly the counter-example here. The 'code' was the easy part. The moat was the decade of hardware R&D and the leverage to force banks to adopt a new standard. An LLM might write the API wrapper in seconds, but it can't hallucinate a relationship with Visa and Mastercard. You literally cannot create a new Apple Pay in a week or even years, no matter how much you vibe code.
I'm sometimes baffled by what people think can pass as a product in a real sense.
To the extent that’s true it has much, much more to do with AWS, open source libraries, and collective knowledge, than it does LLMs.
But I honestly can’t think of anything you could do in a week that a company in 2015 would have paid billions for unless it’s something like tweaking an LLM. But in that case it’s the original model, not the 1 week or work you put in.
I mean literally anything that leverages modern APIs.
WYSIWG Site Builders, text Chat bots, audio Transcription, voice synthesis.
Yeah building from scratch would take longer, but you can slap a UI, a DB/schema around modern APIs and output something that would be science fiction 10 years ago.
WYSIWYG sure builders existed 10 years ago and did not cost billions. Chat bots were a novelty since the NFT bubble hadn't popped yet, and they would invent NFTs to stake the economy on instead. Audio transcription and synthesis existed and did not cost billions.
It's just another way of saying "Ideas are worthless, execution is what matters" which has always been largely true.
Yes, you need the idea first of course. But that's truly the easy part. 99% of "ideas" rely on great execution to be worth even looking at - much less paying for - for anyone else.
That isn't what it's saying and I don't think the idea that "execution is what matters" is even true, other than to point out that ideas aren't valuable by themselves.
This is about marketing, about getting people to know and care that the thing you built exists. You can execute perfectly (in terms of making a great product) and not get a single eyeball.
> You can execute perfectly (in terms of making a great product) and not get a single eyeball.
that's a tautological statement - if not a single eyeball is on the product, then you obviously didnt make a great product. after all, who determines a product is great? It's those eyeballs, not the creator.
You've basically chosen to ignore the whole AI argument as if it was just another tool and we had business as usual. Given how pervasive and fast developing it is, there should be an argument why it can be dismissed so easily.
This AI boom is just a hyper-version of previous tech booms (web 1.0, VC, crypto, etc). You have an enormous number of people who just want to get in and build something, but the products they are pumping out don't serve anyone's need or solve anyone's problem.
The moat isn't money for out-marketing your idea that 750 other people are building, it's having a good idea that solves a problem that nobody else is solving well.
I worked during the digital revolution in film, I've told the story a zillion times on HN but basically, I went through the first pure digital film program in Canada, by the time I graduated 70% had dropped out, as far as I know I'm the only one who made a proper go at it, and even then when my startup was taking off, a new hot shot would show up every month and be gone the next when they got bored or frustrated when nobody thought they were special. Tools are tools.
They often WERE creative geniuses with a digital camera! Brilliant people! In fact they were the ones who came and went the fastest, what makes you useful in industry is very frequently nothing to do with genius or creativity or mastery of the tools. It's things like reliability, stability, team spirit, open mindedness etc.
Creative genius is one game, but it's far from the only, or even, main one.
yeah this is really a part of it. Both founders and investors get spooked by the rapid entries into a market but persistent not just when it’s hard but when it’s boring goes so much farther.
> the products they are pumping out don't serve anyone's need or solve anyone's problem.
This isn't true though.
Yes, there are too many products being build that don't serve anyone's needs or solve anyone's problems.
However, many of the AI products do solve problems and serve needs.
You're right though, to compare this to other booms, which also had the same problem. This is very much a "hyper" version, which is pretty incredible to be in the middle of.
I don't mean that _all_ AI built stuff is useless, just that the number of products where 'marketing budget' is the bottleneck is dwarfed by the number of tools that aren't that special in the first place.
If you have a product that:
1. Solves a real problem people would pay for
2. Is not trivially replicable by your potential customers or competitors
3. Does not have a natural discovery mechanism by potential customers
I think item #2 in your list is the real kicker here. Given that AI can write code the threshold for "trivially replicable" is going down.
Unless your thing has strong network effects or a large capex requirement (ex: GPU infra) its easily replicated and I think that's really what makes things hard.
Ok but pick any category of human endeavor and 80% of it is garbage in the beginning. There were 3000 car companies in the 1920s, and most of them sucked, and so they died. The market over time will sort out who survives and who does not.
It will take a few years for investors to figure this out, but in the meantime, everyone is spreading their bets around like peanut butter in order to be in the game.
> However, many of the AI products do solve problems and serve needs.
Every solution to a problem comes with its own costs. It is entirely possible that most solutions that are rooted in modern computing technology have actual or perceived costs that exceed the value of "solving the problem".
The problems that most people have that they really want to solve are not addressable by AI, or computers, or software.
This seems to me like the few booms I’ve seen before. Absolutely crazy valuations with very little behind them, massive hype, everyone’s unemployed uncle suddenly becoming a shallow expert. It’s probably going to end the same way too, once the upward momentum dissipates and things start to retreat to “fundamentals”, we’ll find out that there were a lot fewer solid points in the market than we were all told to expect, so the fundamentals are actually pretty far down. After 5 to 10 years of regrouping, a more mature and solid version will come about and become such a normal part of life we barely even remember what it was like without it.
We are well on our way to the popping of inflated expectations.
Currently people are taking AI hype too seriously and extrapolating its success out in such a way as to discount the value of other businesses.
Example - last week a bunch of trucking stocks crashed 10-20% because a $6M company that pivoted from Karaoke to AI demoed something.
This is just insane. Sure, if say Waymo is pivoting into commercial trucking.. maybe. But people are basically shorting minutemaid lemonade because their neighbors kids opened up a lemonade stand. Demos are easy, products are hard.
> …it's having a good idea that solves a problem that nobody else is solving well.
Added emphasis to the most crucial part, in my opinion.
If you can manage to deliver a product that's meaningfully better than the competition, you still have an edge, so long as you're competent at marketing.
Nothing gets people searching for alternatives as consistently as frustration, and a product that was lazily built with AI (vibe coded or otherwise) is going to be full of bugs and papercuts that make using it a poor experience.
This is particularly true for software that sits in the hot path of peoples' workflows, where thoughtless design, misbehavior, and poor optimization chip away at time the user can't afford to spare.
In short: yes, competition will be plentiful but it will also be almost entirely awful, and capable SWEs can capitalize on that. It won't take much to stand out amongst the mountains of garbage that will be generated in the coming years.
> The moat isn't money for out-marketing your idea that 750 other people are building, it's having a good idea that solves a problem that nobody else is solving well.
This is a very naive take. Any good idea you have can now be cloned trivial in no time at all.
The clones just need to out spend you on marketing even though it is your idea that the LLM cloned.
Which is my point: it seems easy to clone something that seems conceptually simple, but is the result of all kinds of UI, UX, performance, etc optimizations. The reason someone might choose Obsidian over these so called clones is not just marketing, I assure you. The reason people attempt Obsidian clones is that they think creating todo-list management tools, etc is all it takes to implement a PKM, and that that is easy. It is not.
Right now there are kinds of tools I wished existed, that I would pay for, but AI does not automatically provide the insight, good taste, technical excellence, and grit needed to create these products. I could do them, with AI assisting, but do not have the time. It is not a simple matter of saying: Claude, create or clone X, Y, Z.
There are so many of them, aren't there? There's Roam Research (might be the OG one), Logseq (FOSS Obsidian basically), Notion, Emacs' Org-Roam, Anytype, etc. Neovim has like 5 extensions implementing the same idea (such as Neorg), Bram's Vim probably has its own plugins in Vim9script.
> Just on HN alone, throughout 2023 - 2025 we were seing like one new TODO app show up on HN weekly!
This response shows you missed my point entirely. I am saying a todo app does not a PKM make! I'm not interested in a vibe-coded todo app, it is useless to me.
> Because money is the moat, and they have it!
You're sidestepping issue and contradicting yourself. I asked: if cloning is so easy, why has no one cloned the JetBrains IDEs, for example?
Remember, I'm not talking about what happens after the cloning. Are you saying no one has cloned Jetbrains because it takes a lot of money to do so? That would contradict your claim that AI makes it easy.
What's the point you are trying to make here? That it is not possible to trivially clone good ideas?
Let's start with my actual claim - "It is now trivially possible for someone to clone your good idea".
I want to clarify your position: Do you think that the bar for cloning someone else's good idea is now:
1. Harder to do with AI,
2. Easier to do with AI,
3. Exactly the same level of difficulty it always was.
Because if you are arguing that #2 is an incorrect answer, there's no real point in continuing the argument, is there? I'm taking #2 as a given, and you appear to be arguing that it is a baseless assumption.
> The moat isn't money for out-marketing your idea that 750 other people are building, it's having a good idea that solves a problem that nobody else is solving well.
An idea is not a moat. Execution is only a moat if being nimble is part of the ongoing offering.
Historically, during booms like this (for example the industrial revolution), it was, because we had patent protections in order to encourage ideas to be brought to market. I don't see how you can have an AI revolution that doesn't just funnel everything to the top without something similar.
Why invent the cotton gin, find investors, and bring it to market if the steel company with the infinite worker machine can instantly compete with you?
The discussion here is going sideways, and I blame the underwhelming blog post.
Having money is NOT an economic moat-- i.e., a durable, structural competitive advantage.
He overlooks broader, true definition of moat attributes like labor supply, infrastructure, PP&E, brand, network, natural monopolies, switching costs, regulation. These don't go away with commoditized CRUD apps.
And quoting someone with decades of experience implying that things are hard now and innovation didn't turn over industries in the last 25+ years is a joke.
The more I think about it the more brainrot the article really is. As if all problems are solved and pumping out soulless shovelware companies is worth anything. It really is "just one more app" all over again.
Knowledge ? For b2c it might be more difficult, but in b2b, understanding your customer and their specifics issue and developing something made for them is one of the big challenge. Being able to spit out code for free is useless if you don't know what and who you are making the code for.
You work on niches that have very specific requirements that you can only derive from having a good relationship with customers and so you attend to those needs faster than competitors who are out of the loop.
In addition to that, what they don’t mention is that:
1. Other app stores like Google Play and Steam haven’t seen this rapid rise.
2. There are thousands maybe tens of thousands of apps that are just wrappers calling OpenAI APIs or similar low effort AI apps making up a large percentage of this increase.
3. There are billions of dollars pouring into AI startups and many of them launch an iOS app.
Has steam not seen a rapid rise in AI-asset shovelware?
I'm not talking about the AAA or the AA or even the A space (where AI is being incorporated into dev processes with various degrees of both success and low effort slop), I'm talking about the actual bottom of the barrel.
You never needed AI to make shovelware, you have been able to make a shitty game over a weekend ever since RPG maker was made and there are still games made using that.
AI just helps create some assets for games, it doesn't really make it easier or faster to make games but they might look a bit better.
I can’t speak to the quality of all the games released, but in January 2025 there were 1,413 games released on Steam and in January of this year there were 1,448.
> It's like that FT chart claiming that the rapid rise in iOS apps is evidence of an AI-fueled productivity boom.
I mean, there is evidence for some change. Personally, I'm sceptical of what this will amount to, but prior to EOY 2025, there really wasn't any evidence for an app/service boom, and now there's weak evidence, which is better than none.
Because so much technical functionality has been lost/paywalled/dark patterned/enshitified, I've cut the number of apps I use. I've realized building core personal functionality around the whims of corporations eventually just gets weaponized against me, so I might as well start undoing that on my own terms. Who in 2026 is really bringing in a new app/Saas to do much of anything like we naively did a decade ago? No one I know, we've been shown we will be treated as suckers for doing that.
Well, if someone could create a drop-in replacement for Oracle-DB, Adobe Creative or AutoCad in a month of vibe-coding, I think they could prosper without having their own good idea. I would like to see something like this.
But if you're just talking the fairly simple apps that a single very talented programmer could pump out before this, sure, yeah.
Eh, there's some truth it both. The truth is somewhere in the middle of the spectrum. Distribution absolutely matters, often times even more than the product. And vice versa
No, the moat is having a seemingly good "scalable" idea that solves a problem a few people have and most people think they have. Then getting bought out.
Not really. It's money/resources. I had some really useful apps I built for myself. I looked at releasing them, but there were companies whose business model was waiting for people like me to release apps that solve a problem, and then just instantly jump on creating a solution and outcompeting the independent app.
It's trivial for competitors with bigger pockets to outcompete you on your idea, and there are companies whose business model is just that. And with AI customers are trying to do it themselves as well. The only startup I wouldn't be nervous about as a small team without large financial backing would be ones where we start out partnered to multiple companies in the targeted industry so that we can leverage that connection.
Historically this is why we have copyright/patent laws. To make it make sense for people to try to bring their ideas to the world. But with everything changing we are back to everyone just sitting on their concepts/solutions unless they have big money behind them.
In Republic I, Socrates distinguishes the art of medicine from the art of wage-earning. One is about the work; the other is about getting paid. Historically, the craft was the primary goal, and the money was an extrinsic side effect.
Today, the money-making side has staged a hostile takeover.
The attention conundrum is just a symptom of a deeper financialization. Multi-billion dollar companies have turned profit into a data-driven science – analytically turning the screws on every script, product, and interaction to optimize for extraction. This is the destruction of the art of making things.
The real issue is that you cannot compete with an entity that has no respect for the art. When a platform replaces the integrity of the work with the logic of a metric, the independent creator is no longer an underdog – they are functionally excluded. You can be the best at any art, but in a system that prioritizes sheer extraction over excellence, your craft effectively ceases to exist.
It seems that we have forgotten how to distinguish between value and profit, and now celebrate the latter instead of the former. Currency enables ever broader and more niche markets, but the financialization of everything is the Faustian bargain; we gained niche hobbies but lost our souls?
Not only value and profit, but also assets, cash flow and debt.
Altogether in terms of currency, very hard to distinguish in the most meaningful way.
It's too easy for the real non-dollar value element to slip through the cracks, and even end up completely gone, in a system where true value itself is not prominent enough to be recognized as the source of highest growth, like it often happens so many times but has been forgotten.
Success can spoil you too.
Once you've got so much more cash than value at any one time because you've been leveraging a productive opportunity well, and taking profits aside regularly, if you're not careful the currency amount alone can seem like enough on its own.
Luck has always been a solution to reach. It doesn't scale, though.
In the case of OpenClaw I think you're looking at a fairly pure iteration of luck there, too. It isn't even a case of "I prepared for years until luck finally knocked" or any variant like that. It was just luck.
If that is the only counterexample I'd say it doesn't disprove the point, if anything it just strengthens it. Nobody can build a business plan based on "I plan to be as lucky as OpenClaw".
like another guy said, it's rhe marketing thing. the open claw thing became a hit sensation in the news etc. and now OpenAI can claim it as theirs. they have infinite money anyways, so might as well buy stuff like that I guess.
It was 100% the brand. The creativity may be a bonus for OpenAI. They saw a shitload of people buying Mac Minis specifically for this software.
I don't find a single counter-example compelling. I guess as evidence that "only moat" is somewhat hyperbolic?
But to counter the counter-example, what would have happened if he did not join? OpenAI can just write and release their own version. They can then do the typical loss-leader and advertising tricks that OpenClaw cannot.
The "simply write and release" is what used to be a barrier.
He was competition. They can't have people running their own self-hosted LLMs on their own hardware, and realizing it AI be useful without a subscription!
I disagree. I think creativity is still a valid moat. You still need to build good products. Its like a restaurant anyone with some money can open a restaurant but you need to has the creativity to make a good one.
The only moat that can beat the money moat, in fact
Money (better thought of as credit, since we are talking about fiat here) is an attractor so much as it can stand for or purchase productive energy. If that fails (central bank failure, currency failure, government failure), creativity takes its place
You don't need a ton of creativity for a good restaurant if you have good staff, and upkeep. I'll take a boring well maintained and staffed restaurant over an overpriced "creative" restaurant where the waiters are terrible and the chef is even worse.
This assumes the market is not oversaturated. It does not work if there are only a couple of people trying to find a place to eat in a street with 30 similar well maintained and staffed restaurants.
Correct me if I'm wrong but, you may be thinking of restaurants where their defining factor is "having a very creative atmosphere", which will not suit all customers. However it is a differentiating factor which will serve a big audience that is under served in a location filled with "boring well maintained and staffed restaurant".
In my view the creativity comes in in finding solutions to a problem, in a oversaturated market the problem may be "how do I persuade customers to come to my restaurant instead of my competitor?". And following that question may be (in the restaurant example) "what can I offer that is under served in the current market?" The solution to that may be "a biker cafe" or "an overpriced "creative" restaurant where the waiters are terrible and the chef is even worse" (perhaps even rude on purpose).
Additionally I assume you want the restaurant to grow. If you want the restaurant to just survive the bar is lower and you may be able to do that by doing the same everyone else is doing as long as you meet the minimum.
But the argument is that now you have AI for chef, staff and upkeep so its only the recipe writer left. I don't agree with that argument but that is the dream these corporations are selling.
I can't remember the last time I ate anywhere because of an ad. I eat out a lot too, used to uber eats nearly every single day. I wont reorder at places that mess up my order, but consistent quality I reward. Chick Fil A gives me the best service out of any fast-food restaurant, so they earn my repeat business.
You don't even see ads for local restaurants, you just go there because its close or a friend recommended it. How could you get influenced by something that doesn't exist? Their marketing is just the storefront, people go there since they are interested in this new place.
They definitely spend on ads but I worked at Red Lobster a few years ago, I definitely remember TV ads. Now and then I get a food ad for a place I'll never go to and never do. Uber Eats is trickier because I just look at the map for whats closest to me.
Google Ads, especially Google Maps, has tons of paid ads for local restaurants. So does Uber Eats and similar ilk.
I particularly despise these ads in Maps because the ad often obscures the search result I'm looking for -- and I end up accidentally clicking on an advertisement for some other restaurant than the one I was looking up.
I distinctively remember when ads have worked on me to buy something. Sometimes I blacklist a brand if an ad is deceptive and makes me click on something, but I don't watch TV much if at all, and the streaming services I do use I pay to have ads removed.
You remember when you remember an ad works on you, of course that is a tautology, you don't remember when you don't realize an ad worked on you but did.
Creativity is always a valid moat by definition. If what used to be creative isn't anymore you need creativity by definition to differentiate yourself from that.
They're the exceptions that prove the rule. For every Taylor Swift, there are probably tens of thousands of songwriters who have never earned a dollar from their creations.
The vast majority of valuable copyrighted/patented creations are works for hire, which is compensated just like other labor. Or the creator signs or licenses the rights away to a publisher and maybe receives contractual royalties, rarely more than the initial advance.
(An alternative answer is "the pay for your creativity is a capital asset. That asset can earn royalties. Usually it doesn't.")
I think creativity is still a valid moat. And I think creativity will become a very important moat!
I am a firm believer of "limitation breeds creativity" and being able to make what you want with immediate result will lead to less diverse solutions. In a world of in-diverse products you need more creativity to stand out as it will become harder to conceive of something "outside of the box".
Similarly to how anyone can basically create any image they desire in Photoshop (with some limited training). Leading to a lot of images of a similar style instead of a lot of different styles (ai slop anyone?). This is because between the idea and the result the roadblocks are reduced, the process is smooth (tools aim for 1:1 conversion). In the creative process these roadblocks are usually where you will find interesting new directions or ideas. And very often the original idea was not that interesting to begin with (and perhaps not as original as we would like to believe).
The claude.ai web UI has a bug right now. If you inline some code by typing an opening backtick then the closing backtick swallows your space, puts the next letter you type after the code, then everything else back inside the code span.
One day we might be able to write software without bugs. That day is clearly not here yet.
(Firefox in Linux if anyone wants to repro. Can’t file a bug as it’s a closed, proprietary piece of software.)
That's my hope, too, since I've had the unfortunate luck of practically being forced to fulfill my lifelong dream and create a software company as a solo dev in 2025. Great timing...
However, I've gotten doubts about the value of creativity and having a vision. Plenty of people have great creativity and a fantastic vision, and we can see already how AI allows people to churn out interesting and useful projects at much faster pace than just a few years ago. The problem is that it seems more and more impossible to actually make a living of that.
Anybody with the money to have access to enough AI will be able to create a clone of software X. We might not be there quite yet but it's just a matter of time. That seems to be a death sentence for any small software business.
I think, in time, it will be shown to be the real, persistent differentiator. And far more valuable.
But it's also not a moat in the same way. It's accessible to everyone, but you have to actually disregard the parts of yourself that want to drive hard in some direction just for money or power or external validation.
From the look of things right now, it may take some pain before that really gets to shine.
This article sounds like "I can't make a profitable SaaS, hence others can't too, and those who can, are using absurd amount of money or prior reputation to do so"
I advice to grow up and do some proper research on what problem to solve, and how to build trust with audience before pushing your products.
I dont have much money or prior reputation when starting out, I blogged weekly, then slowly launched technical ebooks, one time utility app, and finally SaaS , took me about 6 years of showing up every day to have a job-replacing income.
Your other post mentioned you gave up and declared failure after 3 weeks of launching your SaaS, I think the timeframe expectation you had might be unrealistic.
I was there in the 90s. I built a few bad static HTML pages. It wasn’t hard. There are lots of stories of non-CS / non-technical people making stuff from the dotcom era.
Making a dynamic page was harder. Integrating with a payment system was almost magical; there’s a reason PayPal became big.
But what was truly hard, and continues to be hard, is building a page, either static or dynamic, that people actually want to visit.
Building static pages that worked in both Netscape and IE 4, and could function well with constrained dial up speeds, may not have been "hard", but it did come with a number of challenges.
Ahhh that lines up with a thought I had recently: you get out of life what you put into it. I am beginning to believe that there's some kind of metaphysical rule here that is true everywhere, all the time.
So maybe the solution is: find the hardest stuff to do and do the crap out of it.
The entire pitch of LLM companies is that that’s not true. The LLM does the hard work, and you pay the LLM company for the tokens. So the gate is just, can you afford enough tokens.
Not saying that’s what will happen in reality, but that is the marketing pitch
Coding agents are good, but once the complexity is high they're not good enough. Eventually your agent won't be able to make changes to your code base without introducing bugs with every change. In my experience, the agents aren't very good with abstractions yet, and no amount of testing can completely paper over that problem. So yes, the industry is changing dramatically and at a breathtaking rate, but I don't think money is the only moat left.
Not just complexity, but also anything requiring any actual ingenuity. It's impressive how "junior" it can be, with its (current) statistical shackles. Worse, if you flat tell it what to do, if your approach is too far into the statistical weeds, it'll flounder like crazy.
Just tonight I was trying to get Claude Opus 4.6 to design a lock-less data structure, and it kept failing to solve problems which I kept then solving for it, but once presented with the solutions it was able to check them, and I think correctly so (that's not nothing!).
And let's be honest that's not a new thing. It's been already a long time since you had a revolutionary idea in the shower only to google it(or use an LLM nowadays) and discover that there are already eight different apps that do what you were thinking.
The way to tell if a business has a moat that I once learned was, "If someone gave you a billion dollars, could you go compete with that business and have a reasonable chance of winning." If the answer is yes, then the business has no moat. The numbers are bigger now, but I think the principle remains the same: money cannot be a moat.
A moat is around something that exists, meant to protect it. You're describing something else, not a moat with your example. OP used moat correctly, creative effort around existing earned skill, brand, etc.
To answer your question. Yes to a player already in a market with lack of funding, a billion dolars could be the necessary moat to win.
Money was, is and likely will be a moat for a while. But as a proxy, it may not be enough as a moat. Scarce resources may require more than money— e.g. IP classes or, if you're China, ASML machines.
These types of takes are getting really tiring. The game isn't over, just the metagame.
I'm not saying I have the next metagame figured out, but AI doesn't magically solve all the worlds problems by existing. It likely even creates a new classes of problems. In such an environment there's going to be opportunity.
Are the software shops and devs who were winning at churning out SaaS products going to be the best suited to this new environment? Maybe not, but adaptation wins over non-adaptation, something that has always been the case.
The well is drying. You have less money available for hustling and less small players with money. With layoffs happening everywhere you have more people with ambition and time on their hands, but less of them can afford big expenses or major risks.
The hype machine currently pushing for agents is selling agents ability to do automated marketing. However the bigger companies know better than to create giant security holes and the small players are either not technically skilled, or will balk at the huge per-use fees for the good models, or will be drowned out because of low quality cheap model output.
I had this conspiratorial thought, would a model really just spit out some money-making scheme, or would it be blocked. Made me think to run the model myself on my own GPUs but still a black box, at least you control your prompts/data flow locally.
The OP should read up on business strategy, probably starting with Porter's Five Forces.
Hint: Money is a nice thing to have but it is very definitely not a moat.
If a company is making above market returns and the only thing stopping a potential competitor from competing with them (aka the company's so-called "moat") is that "it takes money", that company does NOT have a moat.
It's very easy for a potential competitor to calculate that, after x-months or y-years, they will have made enough profits to pay for the cost of building the competing product. As long as that amount of time is finite, there is excess profit for a competitor to take, and the company will find it's so called "moat" wasn't a moat at all.
This isn't a new thing. It's been a fact for centuries or millenia. It's one of the many things that makes success in business hard.
Porter's Five Forces is one distillation of the foundational principles on which moats can be built (and yes, this is a non-trivial subject, so success in this area generally does take more effort than just reading or skimming the Wikipedia page, but if you had to distill it down to one sentence, it's probably "try to build something that has network effects").
> Hint: Money is a nice thing to have but it is very definitely not a moat.
in this context i think where money becomes a moat is when you consider Google can keep advancing their LLMs and tools while operating at a loss because they have the cash and revenue to do that. They'll just keep going and wait for everyone else to go bankrupt and then the whole AI market is theirs. In that way money, and the ability to out last your competitors, is the moat.
This person needs to study Ronald Coase's Theory of the Firm in order to not make incorrect assertions like 'the only moat left is money'. There's many other criticisms to level at this piece, but this is the one I'll use my reply/comment on. If the author was right, then the entities with the most money, or ability to print money, a.k.a. the government, would be in charge of all economic activity. They are not. Empirically, what we observe about human economic activity is that it's not about who has the most money. Money helps, and certainly there's economic structures that are a tier below government but definitely far larger than a small business that can use money as a weapon to stay entrenched or gain other advantages in the market, etc, but it's not the only thing. Empirically it is just not. There has been, is, and will continue to be room for what other commenters have called 'creativity' broadly speaking, and that even includes creatively recruiting other people to your cause/company versus another's, etc, that 'the one with the most money' has not alone shown to be decisive.
> The people winning mostly had a head start. Or they have money. Usually both.
It feels like that doesn't it? But, as one counter-point, OpenClaw. :)
Btw I did a deep-dive into AI moats last week and wrote a blog post about it. Relationships were most likely the strongest moat from my research - but definitely having a large amount of money in reserves helps. https://www.cjroth.com/blog/2026-02-11-moats-in-the-age-of-a...
This is not new. When I started my first company in 2012 it was bootstrapped and getting anyone to pay attention to what I was building was almost impossible. I had to pound the pavement and meet people in person at coffee shops and pitch to get my first few users.
Then when I raised from a16z and had some money in the bank it didn't get any easier. The money didn't help (maybe it wasn't enough). Ad spend or content marketing or paid channels were all hard regardless of the free vs paid.
Maybe I just wasn't good at it.
That was before AI and you had to manually pound the bits into place.
Now with AI yes there are a lot of people shipping a lot of things but humans can tell when someone's put effort into something vs not and the time to traction is still as high as it always was.
Someone should do some analysis on number of things that go "viral" or gain adoption quickly today vs 5 / 10 years ago.
Getting traction has always been hard. Thats just business.
This is certainly one conclusion that could be drawn.
Another conclusion could be that as building software gets easier (like it did for ex in the 90s and again in the 2010s), opportunities are created for new entrants to displace Bad Old Software.
Those expensive Enterprise apps that everybody hates? Are absolutely begging to be replaced by something better for half the money.
We still live in a world where most individuals own more compute power than most universities did in the '80s, yet the only sign of automation is useless push notifications.
Data behind one pane of glass can't easily be moved to data behind a second pane of glass. Simple stuff like "move my Instacart shopping cart to Costco.com same-day" is a manual affair. This is a subset of the general problem that more apps has resulted in more data silos that are generally isolated, without APIs, without automation.
There are zillions of problems out there for which people will pay money, but money chases the same 4-5 problems at a time. Just work on one of the other ones.
I've been enjoying smashing through those panes of glass with LLMs. Just spin up a browser, go and export all the requests to undocumented user endpoints to curl commands, specify what you want out of it and after a few minutes you have an api endpoint up and running!
> as building software gets easier [...] new entrants to displace Bad Old Software
This didn't happen for music.
It is much easier to create/record music today than in the 70s and 80s, but the music created today is mostly boring AI music and not new exiting/inventing music.
The value of human repeating other human's know-how through hard work and thinking is going down. Unfortunately (or fortunately?), many types of programming is one human repeating other human's know-how.
I used to agree completely, and a part of me still does. But the part of me that's scared that this isn't true is getting louder and louder.
I'm a novelist and software engineer. The value of one of those skills is trending to zero. I'm not seeing much to suggest, in the face of the hockey stick of doom, that the other isn't.
If anyone can create what they want won't coming up with good ideas be the value, therefore the creativity will become more important?
I don't know what you wrote and I agree that it is very sad that the skill of writing is very undervalued at this moment. But I think that trend is not new and in reality not created by AI (just accelerated). But I think value of writing was never in the writing itself but the knowledge and ideas behind the writing. Because mechanics of writing sped up (handwritten, typewriter, word processor). But the process of writing a novel did not speed up significantly as the bottleneck is the ideas and iteration on those ideas not the delivery of the output. Similar to how programming is more then typing code.
So I think the real crisis right now is that people equate typing speed with programming speed (or writing speed) and due to advances in the former undervalue the later.
Sadly the reality is that it won't pay for the bills anymore.
Perhaps in a while people will realize how important the quality of ideas (and the iteration to arrive at them) is.
Sometimes what needs translated to food and shelter isn't your thoughts, however. If you nurture and value thoughts regardless, this point will be understandable.
> Wrong. Creativity, innovation, intuition, taste - all forms of thought solely inherent to humans, all going up.
How do you define value?
I mean, if the demand remains the same and the supply goes stratospheric, the value has to go down.
After all, purchasers don't very well care that your product has creativity, etc and that that other identical product was just an AI cloning your product.
> a paid, invite-only social network where every person is verified human and there's no algorithm
This seems like an incredibly niche product that only a handful of people are interested in to begin with. It isn't an notable or surprising result that building it resulted in little interest from general audiences.
At the same time, I see the appeal. I feel like 10% of the comments I read lately are "is this an AI response?" - would be nice to be free of that. Probably not possible tho.
I think the other moat is access to non-public data. If you can train, measure, or make decisions based on specific data that the vibecoder trying to clone you can't get, you can keep ahead.
People are flooded by new projects and assume (rightly) that most are low-signal, so they don’t engage. Because there’s low engagement, new projects get even less visibility. That reinforces the belief that nothing interesting can be built anymore.
I ship earlier now (often free and open source) to learn faster, but it doesn’t change the attention dynamics much.
The bottleneck isn’t building. it’s distribution and who already has an audience. Now dont get me started on getting an audience. thats a whole different pain.
I commented this earlier and it was instantly downvoted for some reason, but I totally agree with both points; I also think they are intrinsically connected in many ways because using customer-development to understand and build an initial audience often helps identify distribution channels. Here are a few examples I have found
1) I made a product to help people manage their diets; crowded space so it was hard to find users, following Steve Blank's customer-development process I found that the product should have been made for Dietitians not for end users. This may have been obvious to most but it was but to me at the time.
2) I made a product for Medicare/Medicaid but it was hard to get my foot in the door to even be considered for any of the Gov. contracts, so following the same model again I found that if I used politicians that had made promises during their campaign that my product could help them to fulfill then I could use them as distribution channels.
One thing I disagree with or don't understand is
> ship earlier now (often free and open source) to learn faster, but it doesn’t change the attention dynamics much.
I have found the opposite, it seems like the bar is much higher now and not even the boost in productivity that I have gotten from using AI has allowed me to ship secure/high quality products as fast as I would like to; perhaps I am just too insecure about my work but its all new I guess.
Interesting article! FYI I clicked on joinkith.com and got: "Application error: a server-side exception has occurred while loading joinkith.com (see the server logs for more information)."
* Attention span won't stop them recognizing a solution
* Numerous solutions won't stop them adopting one
* That the developer put little effort into building that solution won't put them off.
The real answer is of course that there's a lot of stuff that being built that doesn't solve people's problems people have, either because it targets a problem that doesn't really exist or because it fails to solve the problem it does target.
Money is not a moat since you can buy money (debt) or sell equity for money.
>The thing I launched last week is called Kith — a paid, invite-only social network
Social networks already is an existing competitive space and making it both paid and invite only obviously will hurt its adoption. I wouldn't have been surprised if this failed to get users even if it was free and even if this was preLLM. A brand new social network doesn't truly solve users problems.
The only real mote was and is mindshare. Build mindshare, don't screw your users, and your competition will have a hard time dethroning you. LLMs will probably force SaaS and software prices to come down, which will be a great opportunity for dethroning reigning champions.
PostgreSQL, for example, has a ton of mindshare. It will be hard to dethrone it.
OpenSolaris was Sun's attempt to recover from the loss of mindshare to Linux. The a company that had become dominant in databases due to mindshare acquired Sun then failed to understand the mindshare play, and it killed OpenSolaris. Explain that one to me.
AI coding assistants make it trivial to switch between models, and developers will switch to a different model from a different provider if the new model is perceived to be better than the current one even if that perception is completely subjective.
Furthermore, the only users who will actually pay for AI are these assistant-using developers and other power users -- people who will not hesitate to switch to something better. We are already used to new model releases being neck in neck and there is no real incumbent/underdog dynamic going on here.
Yes, there is strong mindshare among the normies, who only know about "ChatGPT", but as soon as they graduate into a powe user, then the above applies to them.
I liked the part about attention being the scarce resource now. Everyone is competing for your attention. But then I see a world in which openclaw is managing emails for people and searching the internet for them and shopping in their behalf. How long until we start seeing advertising specifically targeting AI instead of humans?
Not discrediting the post, which I think is worthwhile and generally pointing at the right thing, however:
> I launched something last week. 14 people signed up — no ads, just a couple of posts. 14 real people who didn't have to. That number is tiny and it felt like something. Then I sat down to think about what it would take to grow it and I couldn't look at that math for very long.
They launched a paid social network, with no content available without joining a waitlist.
This would not have worked 20 years ago either. Bootstrapping the content for a _free_ social network is incredibly hard. But a paid social network where the only differentiating factor is that users are humans, and there is no activity in the network? Not going to work.
>as someone who's been building for the internet for 25+ years, this is the first time that i've ever felt like it's very difficult to make money building new things.
I understand it was already the trend 25 years ago, but way before that you really weren't expected to be able to make money building things for the internet.
The internet itself was simply not designed for that to begin with.
Building things where the internet was an element was already getting bad enough.
The force from within to return to "normal" baseline may yield, but probably never go away.
>The people winning mostly had a head start. Or they have money. Usually both.
As said every millennium since institutions and finance have existed.
>Show HN, the one place the internet was supposed to notice if you built something real.
No no no no no. This is for people who want to share with a much more limited audience than the entire internet.
HN readers did notice a lot of times especially when the project is amazing, OTOH sometimes the latest little side project from somebody well-known, or random interest could be shown.
Naturally the most popular things are free since that's inherently the most compatible with the internet anyway.
But real marketing and promotion is supposed to be far away from this site. If you're trying to sell to "the internet" you've got the whole rest of the internet for that.
HN is not supposed to be enough to be widely noticed at all, if you've got something that's worth marketing, YC is there the whole time and might be able to get you making the most of the internet and then some. Especially if you need a moat of money.
But why do so many people think the only business plan is to prepare to be sieged by a small enough horde which can be deterred by a moat anyway?
>if you're not already moving, you might never take off.
>The cost of acting like it isn't true when it is: permanent.
As I first mentioned, the internet being in place so people can make money off of it is the thing that just wasn't true to begin with, lots of people had some pretty good workarounds for a while though.
I've been watching businesses from startups to large corporations lock in high costs the exact same way for decades before the internet ever came around.
For the comment from "Is Show HN dead? No, but it's drowning", the article author editorializes
"One of the great benefits of AI tools, is they allow anyone to build stuff... even if they have no ideas or knowledge.
One of the great drawbacks of AI tools, is they allow anyone to build stuff... even if they have no ideas or knowledge."
into
"One of the great benefits of AI tools is they allow anyone to build stuff, even if they have no ideas or knowledge. One of the great drawbacks is they allow anyone to build stuff."
which removes the rhetorical effectiveness of the comment (and also breaks the promise of a quotation). I recommend that OP represents the source exactly.
____
I now see that this article contains multiple GPT-isms
It would be better if the article editorialized it thusly:
> One of the great benefits as well as one of the great drawbacks of AI tools, is they allow anyone to build stuff... even if they have no ideas or knowledge.
I'd paraphrase it with "said that" since the quotations indeed present as though verbatim.
I disagree. Customer relationships, taste, creativity, tenacity, execution excellence, industry relevance, reputation, non-public data, etc... are all hard earned or intrinsic capabilities that matter a lot more than feature development speed/cost.
I really like this article, particularly the part about the gravitational threshold. My overall thoughts about launching products is that there will just be more products with smaller audiences, similar to how streaming broke down linear cable. As a software entrepreneur, this means going wider not deeper. Use your human skills to build relationships with people who have that gravitational pull. If you don’t have those relationships, go create them IRL. For context, I’ve founded and sold a social network to AOL, helped raise over $100m, and had 2 other acquisitions.
Utilizing connections with influencers seems to be the most effective way to break through the noise now. Ironically, that's how it used to be before marketing and product development became so accessible.
It’s been true for some time. That moat has been getting shallower and shallower everywhere it isn’t inherited wealth. Not entirely shocking, large amounts of human history have been organised that way.
This post was an Ad also for his paid social network, it did reach me because the message resonates with me as a builder but I don't want to sign up to a social network
So the attention was there but not the conversion.
> Creation used to be the scarce thing, the filter. Now attention is. Most of us are on the wrong side of that trade.
"Wrong side" means that people on that side have to work harder to reach us. The terms of the trade did not change because creators are not starting to pay us to get our attention. Well, it happens sometimes: free trials, free tiers, coupons, etc. but it's a well established practice. The post contains a reference to "do more marketing".
"the value of a human eyeball" / attention is and always will be the limited resource. But I wish the way the economy worked wasn't that attention is sold for money, which makes money the moat, and sets a floor on how low-priced things can get for customers too. Is this really the best the economy can do? Or is it possible to have a fair LLM-based search engine that matches customer need description with stated product descriptions from providers (while weighing customer reviews, etc)?
Kith looks interesting, but the main problem there is verifiability. How would it keep bad actors out? How does it verify posts by invitees aren't AI generated?
It does sound interesting and I signed up for the wait list. But I actually don't like pure chronological order. It feels like I have to look at everything to find the good stuff. Here I probably won't miss out too much if I check the front page once a day.
I hope this is true, I'm trying to make a multi-agent orchestration thing that looks at grain futures, satellite imagery, news, etc... to trade crypto. I'll probably lose but yeah.
Maybe the guy doing their 9-5 can run many agents to make them money while they work their day job.
Is that a thing, you get hired at some company then you use an agent to work for you, deep fake video calls, cursor code... that would be crazy. Get another job and split your time between agents for minor corrections.
>Every morning a few thousand people wake up and ship something. A tool, a SaaS, a newsletter, an app that does the thing the other app does but slightly differently. They post it on Hacker News. Nobody clicks.
>This is not new. What's new is the scale. An AI can wake up (or whatever it does at 3am) and ship twelve of these before breakfast.
That's fun, I'm sure if somebody actually checked that and graphed it, you would not be able to pinpoint when AI starts on the graph
Ironically, lots of people are going to get pushed back into agriculture because yields are going to fall and humans can pick marginal/mixed/non-row crops much more efficiently than robots, plus local consumption eliminates logistics costs.
So yeah, techno-feudalism is going to drive us back to agrarian feudalism.
I partly agree.
Yes, the entry level is low.
It helps me to gain time on my life, doing other things.
But I do not believe that much that more people are going in. Most non coder people i know will not code a thing.
Most bad student i got will not start to do side projects.
So, yes people wanting to ship something ship more, but for the rest, there is still place.
> The corollary is rarely mentioned: the value of a human eyeball is going up, because there are only so many of them and there are now infinite things that want to be looked at.
> The value of human thinking is going down. You probably knew this. The corollary is rarely mentioned: the value of a human eyeball is going up, because there are only so many of them and there are now infinite things that want to be looked at.
Hey, so I'm thinking about getting my car washed..
This article reads as overly hyperbolic; cashing in on the AI hysteria. AI derangement.
I like the idea of you https://joinkith.com/ but I don't see how it can possibly work. These are the issues:
There are people who are sick of social media and will never be convinced to join up again. They've already left the building and aren't looking for anything else. I'm not quite there, but almost.
Other people are using established social media simply because that's where their people and orgs publish. I am eternally frustrated when my local cafe uses Insta or FB for their "web presence", but I'm not going to be the one to convince them to use something else. I hate that my local rock climbing partner finder group is located on FB, but what can I do about it? I also think it sucks that there are thousands of people in that group - I soon realized that this group simply doesn't work for me, since rock climbing requires high trust and I can't trust thousands of people.
Many people resist the idea of signing up for yet another social media account, esp when none of their people/orgs are already there. For example, I've sometimes thought of starting a Heylo group for local rock climbers to find partners - this might actually help me find more climbing partners. But I've never tried it. I just don't think people will join. The barrier to entry is (1) install app (2) create login (3) use app. SFAICT no one wants to do this if they're already on FB and already are a member of the group there. Even people that I know manage finding partners with email lists (gag). Can you imagine how much higher the barrier to entry would be if adding (4) you have to pay a monthly fee?
I do like the idea of "only allowed to invite someone that you know in meatspace" but how is this enforced? I also recognize that requiring payment could help increase the trust level, and I recognize that members have to pay in some way (ads, fees, sponsors, privacy violations) in order to support the platform.
> I’m young so that’s probably part of it, only ~12 years into my career and haven’t experienced too many world defining shifts.
I'm old (52) and a bit AI hysterical despite being well aware of the reasons I supposedly shouldn't be (variations of Jevon's Paradox, the fact that we've had similar disruptions before, etc).
I can't help but think that both the speed and massive breadth of the AI disruption across so many industries all at once makes this a very different risk than anything we've experienced before, in my lifetime or before it.
It also doesn't help that at least here in the US this is all occurring when our government is both openly corrupt and particularly dysfunctional at solving any of the real-world problems facing its own citizens.
looks like the real tension here is that "money" and "reach" aren't quite the same moat though. I think the post kind of conflates them. existing audience is the actual barrier - money can buy ads but it can't buy trust or distribution, not quickly anyway
the gravitational threshold thing is real ngl. I've seen the same dynamic in product launches - identical quality, completely different outcomes based on whether you're already above the line or not. that part holds up
not sure the "creativity is the moat" counterargument fully lands either. yeah taste matters, but AstroBen's point is valid - anything that gets traction gets cloned basically immediately now. so creativity gets you first mover advantage for like... a week?
maybe the actual moat is just community? people who already trust you before you ship. which is a form of reach I guess, so kind of proves the point
I think there's still value in building quality products, but AI makes it easy to build something that appears good but doesn't actually work that well. It's very difficult to communicate the thought and intentionality that went into a well-designed product in a way that stands out amongst the noise.
Why are HN mods appending a question mark to the original title which is obviously a headline of a blog post, i.e. a thesis of the author, and not an assertion of fact? You may as well suffix every headline with a question mark.
http is now a trash can no by AIs activity, they only reveal to our eyes something that has been already happening for years, not perpetrated by robots, but something worst: marketing industries. But in other hand, why to pay for another social network when exist technologies like gopher, RSS feeds or gemini-protocol?
There must also be some value in advertising to the "eyeballs" of the AI? Even if they don't (yet) make that much decisions about spending, they do influence human decisions.
It's the new SEO. Enterprise software vendors are already tailoring their marketing pages and docs to influence AI for when some VP asks their chatbot "which XYZ software should we consider buying?" This is something a coworker is facing at work as we try to leverage AI to speed up market discovery/tool selection work.
If you thought things were hard now just wait for the industrial-scale fully automatic fast-follow bots that will nearly-universally nuke the human-created original product to oblivion in a few years...
Your concern is valid, but you are misguided on a few details. I will inform you as best I can in the context of US intellectual property law.
I am not a lawyer, but I have spent a lot of time adjacent to this stuff, and am not aware of a "copyright strike" as a legal concept. I assume you are borrowing this terminology from various platforms who have their own internal "three strikes" policy for user-created content that violates copyright. However, those policies are entirely for-and-within those ecosystems, and do not directly correlate to any legal notion of "copyright strike". If you violate the rights of someone's work, they send you a C&D and/or sue you, regardless of the platform's use of some "strike" system.
That "someone's work" is key: Copyright covers the output of creative works: writing, art, photography, code, music, video production, et cetera. Copyright law would apply if social media Kith stole an image from fashion Kith's website. It would also apply if one was sent social media Kith the source code from fashion Kith's website, and social media Kith used it (I don't know why they would, but it would violate the copy rights of fashion Kith). The violation would be about the work, not the name.
You referred to the brand name, which would be covered by Trademark law. Trademark law concerns itself with more than just the name, but also the design/logo, sound of the name when spoken, etc. Within that, there is a focus on the likelihood that the public would might confused and the degree to which one brand or another might be damaged by that confusion.
This leads to (somewhat common) situations where there are two marks with the same name, but in different industries, because there is no risk of brand confusion. For example "Delta" is a trademark of an airline company, and "Delta" is also a trademark of a faucet company. They have different logos and products/services in different industries (the USPTO has 40-or-so categories), thus there is little risk of brand confusion.
On the other hand, if two businesses have products/services in the same category, even a different name can be a Trademark violation if it (for example) sounds similar enough. Imagine a hypothetical hard drive manufacturer named "StoreTech". If you started a hard drive manufacturing company and named it "StorTek" it's very likely that StoreTech would pursue trademark action against you, even though the name is technically different!
There is A LOT more to Trademark than just this, but these are some factors to keep in mind.
Everyone is excited (or scared) by vibe coded startups, but ask yourself, how many vibecoded apps do you use weekly ? Every single one I've tested looks great at first, but is an UX nightmare with awful performance.
Even worse, my founder friends are all churning slop prototypes with Claude Code with zero product vision and are going nowhere. Is this revolution just a mirage that is FOMO-driven ?
While building has becomes way cheaper (and probably is going to become even cheaper in the future), is building something exceptional really that much cheaper now?
AI has certainly made it so much simpler to just pump "something" out (slop), but did it actually make building something that went through hundreds and thousands of iterations significantly cheaper?
I also like to think AI is really raising the bar for everybody. In the past, you could easily get away launching a product with a crappy landing page and a couple of bugs here and there, is that still the case? Don't people just expect a perfect landing page at this point (when's the last time anybody specifically talked/ thought about responsiveness?) paired with a flawless onboarding etc.?
The author of this article is missing whats really going on:
Hard novel things are still hard and valuable, easy things are now super easy and really aren't valuable anymore.
Lets take an easy thing... like a meal planning app or a todo list or something. Its what we would all call a weekend project, but in 2022 it would be hard to develop, get the legal stuff reviewed, do the marketing, and do all the rest of the operations in a few weeks. That stuff is basically a series of prompts now.
Now lets look at a hard thing, such as decreasing the cost of titanium production (courtesy of orca: https://www.orcasciences.com/articles/there-has-to-be-a-bett...)... to be able to scale operations and bring the cost down you are going to have to do some real science and engineering... LLMs can certainly help you, but this challenge is a far cry from something that an LLM can handle.
*Novel things have just become MUCH MUCH more valuable, slop is way way less valuable*
> When someone suggested the answer was marketing:
> jUsT dO mOrE mArKeTiNg!!!!!
This is a good point. If there's a problem reaching people because the information channel is saturated, the solution is to increase the information? And then everyone reaches the same conclusion and increases.
This destroys the channel. It's not a zero sum game. If everyone markets, nobody will make the sale because the customer will nope out and see nothing.
The value of unoriginal thinking has gone down. Thinking which is quotidian and pedestrian has become even more worthless than it already was.
The value of true, original human thinking has gone up even higher than it ever has been.
Do we think no new companies will ever succeed now? Of course not. Who, then, will succeed? It will be innovators and original thinkers and those with excellent taste.
Why did stripe make big inroads in developer spaces even if they are in an ultra competitive low margin market? They had excellent taste in developer ergonomics. They won big not because they coded well or fast (though I know pc thinks their speed is a big factor, I think he is mostly incorrect on that) but because they had an actual sense of originality and propriety to their approach! And it resonated.
So many other products are similar. You can massively disrupt a space simply by having an original angle on it that nobody else has had. Look at video games! Perhaps the best example of this is how utterly horribly AAA games have been doing, while indie hits produce instantly timeless entries.
And soon this will be the ONLY thing that still differentiates. Artistic propriety, originality, and taste.
(And, of course, the ever-elusive ability to actually execute that I also don't think LLMs will help with.)
This is a compelling assertion. But who among us has truly original thoughts? How much new stuff can there be? If all the same-y stuff is losing value (but most stuff has value because it's FOUND not because it's unique) then isn't net value decreasing?
>The value of unoriginal thinking has gone down. Thinking which is quotidian and pedestrian has become even more worthless than it already was.
Imagine American manufacturing industry workers saying the same thing of the (at the time) soon-to-be import only products. Original thoughts are valued more than non-original ones but maybe, the market doesn't require that many original thoughts to extract max profit...
Saturation of channels due to slop content - doesn't mean that created stuff is not scarce
there's more problems then ever before that need empathetic humans to solve - are you up for the challenge - or you're doing a quick cash grab
due to people using machine gun approaches - spray & pray - we haven't forgotten how scalable human touch is -- yeah at first - you've to do things the manual way - reach out have a conversation - but slowly word spreads around without you spending money on ads | content etc
Yeah 100% there's more legit stuff than ever, but there's also just more stuff than ever. So it becomes a discoverability problem.
To be honest this is what is discouraging me from writing more novels right now. The only reason I'm even considering it is because the love of the craft is hooked firmly in my gullet. Were it not for that I'd drop writing faster than a lava potato...
There's multiple things happening here. People are using LLMs to write, for sure. It's only natural to absorb what you consume, so as people read more LLM generated content they can unintentionally emit it.
And then there's classic confirmation bias; a lot of people wrote in dry academic prose.
I think this was already true but it is becoming a lot more obvious now and the effects of this problem are going to affect a lot more people as we see mass white collar worker unemployment. Some people think with lower costs to make software more software is possible. But you can try to get started building a product only to see an improved AI agent build out your idea cheaply, or a bigger company use capital to copy you and enter the same market, and so on. Also, individuals who are richer can take more risks - the loss of money is not existential for them.
The only real fix is incredibly heavy taxation of income and wealth for the ultra rich individuals and ultra large companies. That is what will break through the money moat and create fair competition.
Is the purpose of space exploration to build rockets? Or is the purpose to deliver things and people into orbit and to other worlds?
Is the purpose of a computer program to use processing, network and memory? Or is it to handle and manipulate information to give results which are useful to people?
Now the moat of having memorized intentionally convoluted and complicated programming languages has been taken away. Exactly like the printing press removed the monopoly on information which was held exclusively by priests and monks.
When tools make a job easier, they open up new markets for people to do and sell things which were too costly to offer before. AI translation alone means that small businesses can open up several markets they didn't have any access to, broadening the number of potential customers immensely.
Everyone now have access to a tool that makes them nearly as powerful as only the most creative builders were before.
But these new builders have a tool, they don’t suddenly have a newfound creativity.
I think with time we will stop seeing what we consider AI slop, simply because we know it’s not worth sharing. Instead great creative people will share very impressive things that simply wasn’t possible to build before.
Difficulty is the only true moat. [Astronaut: always has been]
Current examples: esoteric calculations that are not public knowledge; historical data that you collected and someone else didn't; valuable proprietary data; having good taste; having insider knowledge of a niche industry; making physical things; attracting an audience.
Some things that were recently difficult are now easy, but general perception has not caught up. That means there's arbitrage—you can charge the old prices for creating a web app, but execute it in a day. But this arbitrage will not last forever; we will see downward price pressure on anything that is newly easy. So my advice is: take advantage now.
It was always about money. People never cared about new things or innovation. It was always funneled through gatekeepers. The "moat" was a talking point to VC's who are used to 90% of their ventures failing and yet was taken seriously. We seem to forget all the racist jokes about how China will just make a cheaper copy, where were talks about a "moat" then?
It's like.. what happens when software becomes a solved problem? Like bricks for construction. There's not 50 million typically highly paid brick design engineers in the world.
I'm no big fan of economists - but if some type of business is seen as highly desirable to be in, and has minimal barriers to entry, then the market will soon be saturated. Expecting otherwise is (at best) wishful thinking.
Software to most of this discussion is a web app with a landing page, a pricing tier, and MRR. That's it. The frame is "product," the metric is "traction," and the canvas is "things people pay $9/month for."
But software is instructions that make matter and energy do things they wouldn't otherwise do. It's the most general-purpose tool humans have ever built. So let's actually think about what's underbuilt:
The whole damn physical world is barely instrumented. Agricultural systems, water infrastructure, building envelopes, soil health, local microclimates. There are farmers making irrigation decisions on vibes. Municipal water systems with no real-time leak detection. Buildings hemorrhaging energy because nobody's modeled their thermal behavior. These aren't apps. They're control systems, and they're mostly missing.
Fabrication and manufacturing are being transformed by CNC/3D printing but the software for designing things to be manufactured is still terrible (and inaccessible!). Generative design that accounts for material properties, toolpath constraints, assembly sequences. CAM software is where word processors were in 1985.
Scientific instruments. A spectrometer is mostly software now. So is a radio telescope. So is a seismograph. Every goddamn thing can be a thermometer (accidentaly!) The gap between "data sensor exists" and "useful scientific instrument" is almost entirely software, and most of that software is written by grad students in unmaintained Python.
Preservation. Some people are doing this with datamuseum.dk. But expand it: there are entire musical traditions, oral histories, craft techniques, ecological knowledge systems that exist in living memory and nowhere else. Software for capturing, encoding, and transmitting that knowledge barely exists. Not "an app for recording grandma," but formal knowledge representation of, say, how a master boatbuilder in Kerala selects wood by sound and flex.
Prosthetics and rehabilitation. This one is big for me personally! The gap between what a modern prosthetic limb could do with good software and what it actually does is enormous. Why are my eyeballs still chunks of plastic? Same for cognitive rehabilitation tools, speech therapy systems, physical therapy feedback loops.
Governance and collective decision-making. Every organization above 20 people is making decisions with tools that are basically "email plus meetings plus a shared doc." Formal deliberation systems, preference aggregation, transparent resource allocation. These are hard computer science problems that nobody's building because they don't have obvious MRR.
Tools for thought that aren't note-taking apps with backlinks. Actual reasoning aids. Argument mapping. Assumption tracking. Decision support that makes your thinking better rather than your typing faster.
The entire domain of formal verification applied to things that matter. Bridges, medical devices, voting systems, financial settlement. We have figured out how to prove some? software correct. We almost never do it for the software where correctness actually matters.
And that's me, one person, in five minutes. Every domain expert in the world is sitting on a pile of unsolved problems that software could address, and most of them have never talked to a programmer because programmers are busy building the next task management app.
Go talk to a nurse, a farmer, a building inspector, a food bank logistics coordinator. Ask them what's broken. I promise the answer isn't "nothing" and I promise nobody on ProductHunt is solving it.
PG wrote essays about this ffs! "Make something people want." "Live in the future and build what's missing." That advice didn't stop being true because AI made the building part cheaper. If anything it's more true now, because the building is almost free, which means the noticing is almost the entire game. You are skipping the noticing and going straight to the building, then wondering why nobody cares.
The number of hard things isn't going down. This thread can't see them because it's not looking at the world. It's looking at ProductHunt.
To make things concrete, in the last week I have been working on my open source speech synthesizer, rebuilding Klatt's ideas from the 1980 paper up to modern emotion/prosody work. Did you know the whole field went nuts for neural approaches in ~2018 and there's a whole shitload of interesting papers just sitting out there that nobody has ever implemented in a real system? Did you know that a bunch of people did research into what different human emotions sound like and now I can make a depressed speech synthesizer, or, scarily, one which sounds more honest to people?
One thing I've noticed is that there is a very specific set of technologies that attract a very specific set of people. You may have seen these people jump from Bitcoin to altcoins to Ethereum to NFTs; and now to AI. The common thread behind all of these technologies is that they are inherently dehumanizing: as in, they are a means by which ambulatory piles of money can shed the skin they are ordinarily forced to wear and just exist.
AI seems like the odd man out of the group, until you understand the utter horror that is weaponized post-scarcity economics. "The only moat left is money" is the plan. It was always the plan. The goal of AI - or at least, the goal of AI to the cult of people who mindlessly agree with it is to replace humans with pliant digital slaves.
Power is a zero-sum game. We all live in the same space, exist in the same media ecosystem, and have the same number of hours in the day. The media consumed by the public, the structure of our cities, the investments we make in science, the sources we use for energy, the laws we pass and the way they are interpreted, stewardship of our environment, and the degree to which our tax money is used to drop bombs on civilians are ALL zero-sum.
I am so sick of the pretense that GDP growth means inequality is somehow illusory. Yes, TVs used to be expensive and now they are cheap. Big whoop.
I think you have it backwards. Regulation is what enables monopolies. There is no monopoly for any of the major industries like cow herding or cellular telephone service in Somalia despite almost no effective regulation. There is not even a monopoly of pirates despite them willing to use violence to try and enforce a monopoly.
If you look at the history of the US, for instance, railroad regulation was brought forth largely by the railroads because they found it impossible to form a cartel to keep up prices (due to "secret" discounting) so instead they created regulation that outlawed the kind of discounting that breaks cartels apart. A similar thing happened in banking where the banks asked for a central bank to cartelize the interest ranks to stabilize their oligopoly. And the same in pharma industry -- big pharma loves high regulatory barriers because it keeps competitors out.
A large portion of the regulation in the US was brought about as regulatory capture by corporations to increase the monopolizing effects and destroy the free market.
That's patently false. AT&T was a monopoly and they were broken up by antitrust regulation. The absolute most you can say is that some regulations enable monopoly. I contend that we simply should pass the good kind of regulations instead.
Monopoly is enabled by market forces such as economies of scale. Monopolization is a natural market process which happens on its own unless it is actively prevented.
> big pharma loves high regulatory barriers because it keeps competitors out
The FDA, for all the flaws of its current incarnation, is the archetype of necessary regulation. Pre-FDA, the free market did nothing whatsoever to prevent nauseating practices like the adulteration of milk with powdered plaster, lead, and cow brains. The history there is fun but quite gross.
> Somalia
What is notable about Somalia is not its lack of regulation, but the fact that it is perhaps THE least stable country on the planet. It is not the basis for any useful comparison here.
>That's patently false. AT&T was a monopoly and they were broken up by antitrust regulation.
This is patently false in the context of the reply you have made -- after the invention of the telephone more and more and eventually hundreds of telephone services popped up. Then in 1918 (circa WWI), the government effectively quasi-nationalized AT&T by controlling it via a commission and the postmaster general and then AT&T leveraged politicians to create "universal telephone service" provided by AT&T and regulate competitors out of the market while using regulatory capture to use commissions to regulate rates, effectively creating a cartel that drove competitors out of business via regulation.
the whole idea of a "natural market process" here is absolute and utter hogwash. The majority of the market was AT&T competitor up until the regulators stepped in and turned it into an unnatural monopoly enforced by regulatory capture.
>The FDA, for all the flaws of its current incarnation, is the archetype of necessary regulation. Pre-FDA, the free market did nothing whatsoever to prevent nauseating practices like the adulteration of milk with powdered plaster, lead, and cow brains. The history there is fun but quite gross.
You're now arguing why we need regulation rather than whether they create monopolies or not. I see this as a complete red herring, although an interesting topic, that there are some counterpoints to.
> What is notable about Somalia is not its lack of regulation, but the fact that it is perhaps THE least stable country on the planet. It is not the basis for any useful comparison here.
What is notable is that the whole thesis is without regulation it turns into this monopolized hellscape and every inspection of that theory turns out to be false, and sometimes even the opposite.
Anyway, what regulation is responsible for Walmart and Amazon putting local retailers out of business?
> the government effectively quasi-nationalized AT&T
After a big merger put AT&T in charge of the majority of telephone lines in the US, the company used its control over infrastructure to drive its competitors out of business and increase its portfolio. The Justice Department tried to break up AT&T but failed; it was in the settlement of this case that AT&T was first federally regulated in 1913. Yes, AT&T's monopoly grew between 1913 and 1982, but your causality is backwards. They regulated it because it was already a monopoly.
>Anyway, what regulation is responsible for Walmart and Amazon putting local retailers out of business?
Walmart + Amazon combined are only ~16% of the retail business. They're not monopolies. The fact they put a small minority of businesses out of business does not mean they're a monopoly. This is likely part efficiencies and also part regulatory capture via the insane zoning/building regulations in this country and tax breaks that can favor large corporations.
> After a big merger put AT&T in charge of the majority of telephone lines in the US, the company used its control over infrastructure to drive its competitors out of business and increase its portfolio. The Justice Department tried to break up AT&T but failed; it was in the settlement of this case that AT&T was first federally regulated in 1913. Yes, AT&T's monopoly grew between 1913 and 1982, but your causality is backwards. They regulated it because it was already a monopoly.
... It was not already a monopoly. Hundreds of phone companies emerged and by shortly before your noted date of "regulation" those competitors held the majority market share. It became a "monopoly" after the government literally quasi-nationalized them (AT&T) to the point the fucking Postmaster General was basically in charge of it, they became intertwined with regulators, and then the drive for "universal telephone service" and regulatory commissions ensured the regulatory compliance pushed exactly into AT&Ts business model. AT&T intertwined lawmakers even brought in economics quacks to talk up natural monopolies to argue for policies to create the regulations that made AT&T a monopoly. So you have it backwards -- the regulated it from a minority market holder to an unnatural monopoly and lawmakers created this monopoly under the auspices they essentially needed to legislate a "natural" (misnomer) monopoly into existence.
> Walmart + Amazon combined are only ~16% of the retail business.
of ALL RETAIL? That includes groceries! That's huge! Anyway, Amazon is >40% of e-commerce & Walmart is >10%. Together they control more than half of all online commerce. That's definitely monopolistic.
> It was not already a monopoly. […] shortly before your noted date of "regulation" those competitors held the majority market share
From Wikipedia: "AT&T controlled over 80% of the U.S. telephone system market by 1907 and Theodore Newton Vail rejoined the company as its President. Vail negotiated with competitors, charging them fees for connecting to AT&T's long-distance network. These practices led the Justice Department to attempt to breakup AT&T, but a settlement was reached through the Kingsbury Commitment on December 13, 1913. It brought federal oversight into AT&T and led its Bell System monopoly to become federally regulated."
They had an 80% market share pre-regulation. Yes, it was already a monopoly.
This is not a personal opinion of mine, it is pretty much established science. I think only think-tank backed sources would claim the opposite.
One should understand the phenomenon as a common pattern of dynamics in unregulated markets. Not every snapshot will showcase an end state of monopolist dominated markets.
You bring up a valid point though. Regulatory capture is a indeed a weapon in the hands of anti-competitive players to prevent incumbents. Good policy usually applies differently to different strata: the small players are exempt from certain rules, or have to deal with less stringent ones than big players do, to prevent killing the market. At the risk of sounding like an llm: it is not just about policy, it is about good policy.
The food industry is filled with regional monopolies.
> cellular telephone service in Somalia
Ah yes, excellent example, all you have to do is completely destabilize your nation and you too can have free market capitalism.
Investors love monopolies, they fix prices and profits so their investments are not at risk. Investors hate too much competition, it lowers profits and puts their investments at risk.
Free markets need investors. Investors hate free markets. I hope you see the problem here.
This is a crude but misguided attempt to bypass what I say which was "effective regulation." FGS only controls a minority of Somalia -- most of it is controlled by Somaliland, Puntland, Al-Shabaab, directly by xeer (customary) law, or only FGS hanging on by a thread.
Even in cases where FGS is in control -- even then xeer law on property rights and law often supersede written law of FGS. For instance, on occasion Somali population has straight up killed FGS soldiers/police and Somalian government has deferred to xeer courts and said "welp that is fine." Xeer law in particular has a very liberal free-market view on inter-familial entrepreneurism (although with a lot of intrafamilial and tribal dues which hinder it in practice) which is at odds with what outside law has tried to impose upon it.
NCA has rather limited influence in somalia, and definitely not of the sort that could break up a monopoly if one existed.
There are a lot of zero-sum things that are being chased by money, and maybe even the most sought after ones. Like political power, sexual partners, and chasing the top % of credentials for your offspring to position them with better access to zero sum things.
So, I really felt like more people should be reading Nassim Taleb's Incerto series of books. A lot of the issues that fall out of AI he dealt with in his books like ten years ago.
He gives one the best pieces of advice I've ever heard: if you are going to do something for a living, make sure it is NOT scalable.
If you do something that isn't really scalable, like being a welder or a tailor, then you only have to compete against the tailors in your neighborhood, and you can easily find a neighborhood that doesn't have a tailor. If you're building a scalable product, you'll always be competing against the best, most well funded, smartest people in the room.
Everyone here has grown up in the birth of the internet -- a once in ever event -- where building something scalable was just there for the taking. That's never going to exist again basically.
That seems like reasonable advice until you realize you have no idea what will and will not scale in a few years, and there's only so many tailors/plumbers/welders/etc.
What it really comes down to is these options as more work gets automated:
1) New jobs doing different things that weren't done before
2) Same jobs but shorter hours so "full time" with a salary to match starts to look more like 4 days, or 5 hours/day, or something.
3) Lots of unemployment
These can happen in a lot of different combinations, they can come wrapped up in different ways, and unequally for different segments of the workforce, but there's limited elasticity in most areas where additional people piling into the field would create more demand rather than glut the supply to the supplier's detriment.
I keep hearing about the potential of "new jobs" coming from ai but can anyone actually describe one? My gut says they will be something similar to converting middle class knowledge workers to do DoorDash drivers or trained artists becoming dog groomers. What a cool future, at least my parent's can watch racist ai slop videos on their iPads.
It's simply a case of looking back and deciding this technical revolution is identical to the ones preceding it. Thus jobs destroyed will be replaced with new jobs, it always happens that way.
Of course past performance is no guarantee of future success...
> jobs destroyed will be replaced with new jobs
Not for horses though, or at least not the majority of them. Some were kept as pets or essentially status objects. In this case we are the horses.
95% of human farmers lost their jobs because of industrial revolution. What happened then? No jobs were created and we still have 95% unemployment, right?
And glue, you forgot glue. Maybe we'll get the Matrix plot line where we become human glue. Or, uh, batteries.
> In this case we are the horses.
This is assuming the conclusion. The entire question is whether we are the horses or every other example of humans in the past who found other employment that was inconceivable previous to the technological revolution that rendered their old job irrelevant.
But you are ignoring the substantial difference between previous technical revolutions, this time technology not replacing a mechanical operation.
Horses couldn't revolt.
And we won’t. Uber Eats Burger Reich and all.
The point of doing something non-scalable is that you can enter and exit the market fairly easily. You don't need to be a tailor your whole life. You can make a living as a barber, electrician, teacher, or nurse.
I'm not saying it's easy! It's hard as hell. It sucks when your job gets automated. I'm just saying that aiming for something non-scalable means you're not always tilting at windmills, and the game can't be rigged against you.
Switching jobs from electrician to teacher to nurse will take around 3 to 5 years of education or apprenticeship in most countries. It also requires new licenses or certificates if you ever move country.
I'm trying to respond to this stuff in good faith. Yes, I agree with you. I don't see how this is relevant to the argument. If you are in a non-scalable industry that gets taken over by technology, that sucks. My point was being in the startup game in the first place.
"Becoming a teacher" takes years.
"Becoming a successful scalable business" has no known time frame. It either happens or it doesn't. And whether it does is not particularly correlated to how much time or effort you spend on it.
Almost typed the same thing in different words at the same time :)
Your above point can not be upvoted enough.
You do have to wonder what some people are thinking ;)
Good message from Taleb too, he may not be the most likeable or agreeable person in the world, but he is quite brilliant at synthesizing logical meaning from the inputs he does have.
People not liking Taleb and then dismissing him because he can be rude has always been one of the strangest things that so many people do.
I remember about halfway through Skin in the Game, after being put off by most of what I'd been reading, that the gestalt shift happened and it finally clicked. It's probably one of the best texts I've ever read on a specific type of conservatism. I don't like that I agreed with it, but the guy makes a very, very convincing argument.
Where would we be without Linus anyway ;)
>Switching jobs from electrician to teacher to nurse
The point is not to do that ever.
Just gain one of the skills like that and plant it firmly on your resume, before trying anything more risky.
Like starting a business, which might not actually cost as much or take as long to get a license.
And with no further delay needed before any future pivots, when you might need quick alternatives most.
Even if you only go back to teaching for a while to regroup.
All of those professions you've listed require about half a decade of dedicated training to be legally allowed to practice. For example an electrician takes like 7 years to become qualified, that's a full time apprenticeship, and it pays badly in the meantime.
A fact endlessly annoying to electrical engineers who legally can design their houses power system but not work on it.
(I mean I think a barber is quicker, but one of that list is also not like the others)
There are a lot of code issues you can run into when actually installing and repairing wiring that your average EE wouldn’t know much about at all. And just because you can design the power system for a building doesn’t mean you have any how to fish romex through a wall.
That’s like saying a mechanical engineer should necessarily be able to work as a machinist. Some of them can for sure, but it’s not something they are required to learn.
Also in many places anyone can do electrical work on their own house.
> For example an electrician takes like 7 years to become qualified
Using a similar playbook for what happened to SE (loads of good ~free info, bootcampification) you can drastically reduce the time to be economically useful, but that can only go for so long due to saturation. Especially residential electric work (or other trades) is no witchcraft that would require anywhere near several years of education/practice. And yeah, the legal system would have to cooperate, which will only happen when the pressure becomes high enough (see e.g. debates around "Meisterpflicht" in Germany).
That said if AI continues to progress like it currently does we may not get to that part.
I'm not trying to say 'everything is fine, nothing bad is happening a world of recurring technology and industrial revolutions.' It's not. The way things are set up is bad.
My point is the author writes a column about how GPTs are ruining the ability for people to make scalable products, because when everyone can make one, nobody cares... my point is that that's not the result of GPTs. It's a result of survivorship bias skewing how we look at things.
When your business is a flywheel than needs to be running to provide a benefit to each user, then getting that flywheel running is a huge problem. The vast majority of non-scalable businesses, almost by definition, provide each customer with a benefit regardless of whether anyone else uses it. That is how you create basic, word-of-mouth, free "earned" marketing.
Half a decade is over in a flash. The current software engineering downturn is almost 3 years old now.
Being a tailor is scalable, that's way there are way, way more cheap machine produced clothes today than in the past. Surely he did not miss that the textile industry was at the core of industrial revolution. So being a tailor is more like a post-scaling job - the automation has already happened and now there are only remnants left.
But how can you be sure a job is peak-automated? A few years ago, I would have said musicians are post-scaling - way fewer musicians jobs now that you can play recorded music. But it looks like generative AI will hit musicians again. Can some of welding be automated? Probably.
Welding is being heavily targeted for automation, apart from pressure vessels etc, most welding can be automated now a days , very soon ( months? ) every welding can be automated.
You're thinking of factory welding, manufacturing or maybe repetitive pipeline welding and thinks like that.
It'll be a while until a robot disassembles a trailer enough to remove the bent axle, cleans off all the paint and rust, bends the trailer back into shape where needed, cuts a custom support to makeup for some lost strength, welds it in, primers it, paints it, and assembles the trailer.
Same for construction too.
If automation excels sufficiently, robot will replace bent axle with new one automatically, repair cost will be usually higher than replace cost if all things are automated
Will it tho? I wonder. We're on the brink of a robotic revolution.
I mean a tailor who adjusts clothes and occasionally makes something bespoke.
Tailors typically operate a launderette and act as middlemen to a local dry cleaner.
I’m not talking about a fancy man making clothes for rich people, I’m talking about the talented old lady in you neighborhood who adjusts your clothing for $50 and runs a wash and fold.
How many people visit either kind of business these days? I'm almost 35 and have never once gotten any garment tailored. I think one reason is that clothes are so cheap you can just keep looking and you'll find something that fits, which seems to be what most people do nowadays.
Try it. You might be missing out! One of the benefits of getting all your trousers from a tailor made to your size is you never have to waste time trying on trousers, or waste time ordering-returning items.
Yes, but this is simply the remnants of the old tailor occupation, post automation. The talented old lady would have had a lot more business in clothes making in the past, no need for a wash & fold.
My mom altered clothing when I was younger - and she darned socks too. My M-I-L still sews the occasional seam for pants that are too long and were cut for my wife or daughters.
But me? I buy a pair of $30 jeans at Costco. If they don't fit great, I buy a different pair of $30 jeans. I don't spend $50 to have them altered, or take it to a laundrette. If it can't be washed in our home washer/dryer, I don't buy it. And these days, when a sock gets a hole? I throw it away.
This is an effective strategy if you're a fit model, or close to one. If you're within the standard deviation of the sizing chart, you'll probably do fine most of the time.
I like golf. Most people use a standard shaft. In fact, that shaft length is standard because most people use it. That doesn't mean there isn't an entire industry for "golf fittings" because "most" people isn't even close to everyone.
Well, that's why there are so few people doing this full-time any more.
But at least they're not going to disappear completely.
And now there's nowhere to go but up :)
They have had the internet a while too, keep in mind a select few have gone viral on Etsy while so many SaaS things don't return a fraction of their potential.
>Can some of welding be automated?
Huge amounts have been doing it for decades.
Manual work pays better than ever though.
And plenty of alterations going on all the time after all the automation dust had settled manufacturing most fashions, a lot less manual work is of course being done but it's still everywhere. You do have to be good or you're not going to do half as well as you could though.
The thing is, automation should be expected to slow or stall sooner or later, automation's not suitable for every little bit of welding or sewing that needs to keep going on. Only the most suitable, of course ;)
These are just random examples, if you want to make absolutely sure you won't be automated away by the internet, build a valuable skill that doesn't depend on the internet at all, nor look anywhere near the places where automation is emerging that it wasn't doing before.
If you eventually figure out how to automate that skill it would be something.
Just like the internet though, there can be extra credit for being first :)
One of the most valuable things to be able to build single-handedly is something that can not be mass-produced by any stretch of the imagination.
You might stick with that alone, or pivot to something with more of a financial upside, but you would always have something to fall back on if needed. Plus give you less worry about taking financial risks than you would have been, considering the same resources and/or capital to work with.
And on a regular basis revisit how far you can stretch your imagination to see if your baseline fallback still doesn't look like it will ever be automated in a way that would effect you.
>how can you be sure a job is peak-automated?
Probably the best way is to spend a few years working for a company where you can get a better picture whether it seems that way or not.
I used to look at my local dry cleaner and think that it had to be the most stable business going - it couldn’t be replaced by a computer. Then Covid hit, work from home took off, and the dry cleaner went out of business. Computers came for the dry cleaner, not by cleaning clothes but by eliminating the need for dress codes.
> if you are going to do something for a living, make sure it is NOT scalable
You need to consider both horizontal and vertical scaling. Being a bespoke tailor may not scale vertically, but it can scale horizontally. If you have too many people pick up tailoring, you might run out of neighborhoods without competitors.
> if you are going to do something for a living, make sure it is NOT scalable
I guess it also means that if you build something for a niche audience then big companies will never be interested in it.
That's not really a bad thing, IMHO. Many people successfully make a living creating niche products.
I wouldn't say it's not really a bad thing, I think it's a very good thing. There are many people now making incredibly niche products that have very good lives - making more than enough money for themselves doing interesting work engaging with customers that are passionate about their field.
Sounds like a great life to me.
I'm sure someone else on Etsy will undercut your price if you sell more than a few.
Probably until a copyright vulture comes along and nukes it.
Given the success of copyright cultures in fending off the most blatant, unambiguous case of copyright violation in human history, (AI), I would not put your money there.
I like doing things for big companies that they actually could afford easily, but their costs are so high it's a better deal to have me do it.
One of my favorite niches :)
Some clients are good enough at math to figure it out, and you don't actually need that many of them if they are big enough.
What are the things you do?
One thing: Refresh the page to see the rest of the story ;)
how do you approach big companies to offer them these custom niche services?
Good question.
One early approach when starting to build a reputation is to work as a subcontractor to a less-big company who has already gone through everything to be on the approved-vendor list of the massive corporation.
Even better if the less-big company is a private company and the biggie is public.
A private company will be able to figure out the benefit from your work better than most, and hopefully profit from it.
I like it when every time I invoice a client, they are making money at the same time, whether they are making it from me is not always necessary if I am participating in an overall money-making process they are going through. Especially routinely. It's just fine to be a very small participant in lots of activity going on between the big-shots.
Big companies can have so much work and be able to pay so well that your position can be to merely absorb the overflow from the primary contractor, even if you do not yet have a unique offering.
Then it's good to build into your exclusive-but-related niche, which would be intended for a different division of the biggie and expect it to languish unless your already-established client can use that too.
Eventually after you finally get to meet the right person in the target corporate division, it will be after you have already been doing work for that same corporation through the approved primary contractor.
By this point you've already been doing critical work for that corporation exactly like they need, that you can be proud of and point to, and it can be a whole lot more likely to become an approved vendor yourself. From which position you can finally negotiate fees directly with the most well-heeled corporate source, this would be the first time if you were only working for other primary contractors until then.
Ideally you will then be invoicing a different office of the same corporation that you were subcontracting under, there will be no conflict of interest, and you can continue working for the original primary contractor too.
And your most promising niche finally gets to launch with about as much upside as it can get.
Welding is coming along. Ever heard of Path Robotics? That's high-mix, high-complexity welds. (There are a lot of other, less-sophisticated welding robots out there.) The biggest moat for a welder right now is special certifications to be able to do welds on submarines, etc.
> He gives one the best pieces of advice I've ever heard: if you are going to do something for a living, make sure it is NOT scalable.
You mean like opening a restaurant?
Too bad delivery services like Uber Eats totally own the market now.
Starting a hotel? Booking.com and Airbnb are there to take your profit margins away!
Yes and any other business has to play the Google Maps and SEO game.
I think that delivery services give you bigger market, but not intrinsic scale. You're still limited by kitchen size, staff numbers, and raw hours you can put into the food.
You can scale the system (say Subway, or even smaller chains like Burger Fuel), but also reasonably choose _not_ to scale and still do incredibly well (like Michelin star restaurants, or the myriad of hyper-famous-locally Japanese eateries, or Fergburger in Queenstown).
Someone scaling their own restaurant on the other side of the world won't necessarily out compete out (and in the overwhelming majority of cases with have no impact at all). Despite fast food joints all over the place, I still see small cafes, individual eateries, etc performing well (I mean, as well as hospitality can be).
Maybe it's worth expanding the definition of > make sure it is NOT scalable < to include 'make sure it's not automatable'?
The problem is that any market can be taken over by someone with deep pockets (or investors making a pile of money). They just make sure they are the "go-to place" for consumers to access the market. By marketing the hell out of it, by making apps, abusing power in other markets (see platforms), etc.
That's also the advice of the founder of this website, but not his investment firm since it's looking for moonshots. https://paulgraham.com/ds.html
I'm excited for a future where the technologist is like the tailor in their community. Scaling software has created a host of 'product traps' and there is no need for that for the vast number of activities people do aided by technology.
how good was Taleb at following his own advice? Had he tested it? As I recall he is pretty big on "skin in the game' as his differentiator.
>> you'll always be competing against the best, most well funded, smartest people in the room
What if your goal is for them to buy you out?
There are much less expensive ways of playing the lottery than investing a year of you life into an unprofitable product that must be scaled to be successful, and only then hoping gets bought by a direct competitor.
Don't compete, complement and start building the relationship from the beginning.
Plus, who wants to spend any time competing with the stongest contender anyway, especially to get started :)
Is providing scalable products at a cheaper price scalable? If so, can it hurry up and scale? This is a bit of a paradox.
Who cares? He's arguing about providing something that is not scalable: so whatever happens to things that are scalable ain't the topic.
automation makes every job scalable, no?
If we get AGI and fully autonomous robot assistants, we'll live in a post scarcity world like Star Trek, or somebody in control the robots will use them to enslave all of humanity... so... high variance outcome could go either way.
Do you expect we’ll have AGA, artificial general automation?
I don't think it needs to be general to be scalable to the point that it causes previously safe industries to "race to the bottom"
Uber didn't need to launch in every neighborhood to start destroying the taxi business
Facebook didn't need to have every eyeball on the internet in order to massively disrupt trad advertising
At some point there's just a tipping point
I do. But it will take some time to get there.
I don’t see anything stopping that from happening in the very near future.
could you automate a BJJ coach?
Who's going to pay for a BJJ coach when nearly everyone is out of job?
IMO literally automating the job isn't the only possible scenario -- already in the weightlifting space coaches are supplying skills and system prompts to llms that do their training for them, massively raising the number of students they can train at at time -- at some point that turns training into a zero sum game
I mean, if AI doesn't stall out, maybe there will be BJJ bots in 10 years.
That is the ultimate goal. We are very far from it for many jobs.
Even if it does, there will probably be a prolonged economic period where robots are doing dangerous/messy stuff like welding, plumbing but there is a human master guiding them from a few yards away, via prompts, controllers, etc. More of a semi-autonomous power tool than an fully autonomous master that is delivered by drone on-demand. Scalability is still a ways off.
can you automate a tailor?
Isn't tailors by and large already outcompeted by cheap new clothes?
A big point of seeing a tailor is getting yourself fitted for custom clothing that is specifically made just for you. As someone who's bought $200 off-the-rack suits and $2,000 tailor-made suits, there's a world of difference between the two, especially when you have an atypical body type.
(Granted, to the main point, I still think a tailor could be automated in some distant future, but we'll need robots to perform physical interactions, not just software.)
Tailors are a niche thing for weirdos, now. It's not exactly a growth market. Most folks only wear a suit to weddings and funerals, and maybe job interviews. They have basically no need for more than two suits, and many try to get by with just one (in black, probably). Lots don't own one at all, maybe just a cheap fused-construction blazer or two, if even that. Outright bespoke clothes are a niche of a niche.
Normal people wear clothes containing minimum 2% elastic and perhaps never, ever visit a tailor in their whole lives, except maybe one at a tux rental place or a wedding dress store, for their own wedding. If they repair clothes, it's sewing on the odd button at home or using iron-on denim patches. Past that, it's just not worth fixing, normal folks' clothes are so cheap.
The whole market for tailors is practically an affectation. It's not serving much actual need any more, not from the perspective of the overwhelming majority of people who are happy with stretch-denim jeans and polyester sportswear jackets and such. It's basically 99% of the way to being an obsolete job, kept from total death by a few enthusiasts. Only a bit more lively than the market for, say, authentic regency-era footwear or something like that.
(I am a fellow weirdo, for the record)
I mean, you do you.
Clothing is really _fashion_ for tons of people. Fashion is art. People like art.
As someone who has tried to make several businesses around art, people generally like art but not enough to pay "at scale" money for it.
Yes, I am one of the people who has a preferred tailor who can do more than just let trousers waists out. I also know where the nearest cobbler is. That’s not normal, though.
A dead industry often doesn’t entirely disappear, it just shrinks a bunch and comes to rely entirely on enthusiasts or very rare actual need, rather than broad need or appeal. Consider the draft horse breeder, or the carriage driver. There’s a market for both professions! But they’re itty-bitty. The day-to-day need for both is gone.
Tailoring is hovering right in the edge of that kind of status, today. It’s dying, killed by $10-30 shirts and $20-50 trousers and $50-100 jackets all from largely synthetic materials, and a society that no longer expects anyone to wear anything “fancier” outside certain events.
I mean, outside very unusual circles, dinner jackets are essentially ceremonial costume-wear, and business suits aren’t far behind on that track. You gonna wear a tailored wool hacking jacket or breathable linen Norfolk suit on your camping trip, or a bunch of polyester and nylon stuff from REI? LOL. All the situational tailored clothing but the business suit and blazer are near-extinct unless you want to look like a cosplayer, and those are on borrowed time.
Yes, your message is coming from the pov of economics and business, as makes sense in this thread! That's my mistake, I took your message more sentimentally. I've used tailoring probably 5 times in my life, with the only recurring need being to hem pants.
"There is no money in tailoring" seems right. It's the "not all things need to make maximum $$$" that I speak to. You didn't pick this fight though, I did heh.
My (successful) friend tells me all about how amazing it is to collect very expensive watches. I just need to be a "watch guy" and I'll come to understand. Once my eyes returned from rolling out of my head, I did concede a great point he made: there is no reason for watch makers to exist anymore. The fantastically amazing history and evolution of time-keeping and personal time-pieces is now purely supported by rich people that care to subsidize the art form. And so, maybe I really do aspire to be a watch guy after all... hmm.
A romantic perspective I still try and hold myself, however the point about the watch and the cloth and the dwindling appreciation for such is presently experienced in reference to decades or centuries of disruption and are intrinsically tied to the demand of attention. I don't trust the acceleration will leave much, but I am continuing to paint and taking writing more seriously in great fear of the time scales we are navigating today. I find myself confronted with nihilism in so many facets of my life but perhaps this is simply the smell of the air in my particular milieu.
A huge part of the tailoring business are making small adjustments to cheap clothing to get them 90% of the way to bespoke.
If you’ve never done it, I strongly recommend getting your jackets tailored. Even a casual jacket will fit and look non-trivially better for $50-$100 and an afternoon at your local tailor. You can even get things like cycling gear tailored.
Not if you want any tailoring to be done.
Rich people still get suits custom made specifically to their measurements and preferences. They cost about $20,000 USD. It would be cool to have this process automated and affordable to the masses.
cheap new clothes
Uh.. I don't mean to be that guy, but tailors aren't even operating in that market.
People who use tailors aren't interested in off the rack items.
Even when they do purchase off the rack or even secondhand items..
they'll go have some of those items tailored.
> but tailors aren't even operating in that market.
Not anymore.
Not any time that most of us have even been alive.
It's been well over half a century since tailors operated in the "cheap clothing" market.
The clothes tailors make have, pretty much, always been expensive relative off the rack department store options.
Probably because tailor made clothing doesn't "scale".
Absolutely. Design parametric families of patterns, 3d-scan the person, let customer adjust with live preview, laser cut, then fully automated or low-skill assembly. Probably not currently economical like many things involving physical world manipulation, but without obvious roadblocks.
This seems overly pessimistic
As scalability becomes more accessible like with coding agents
The less it becomes about money but distribution
While money can buy you the latter it may not compound to something that is sustainable
> if you are going to do something for a living, make sure it is NOT scalable.
Great advice but difficult to action though.
I mean 10 years back I'd have thought programming is that thing which is not scalable. I had every reason to believe that. It required skill, experience, ability to stay current, grit for debugging hard stuff. Much of it can be automated now.
What can I pick now for a living that is not scalable today that some future technology would automate it just as easily.
You never wrote any scripts to write code?
> When creation was hard, skill was the differentiator: you had to actually be good to make something worth showing. Now the barrier is near zero, so you need reach. Reach costs money or it costs years. Probably both.
Creation has progressively been getting easier since the invention of the computer, it is not a new phenomena. This naturally pushes the boundary on what needs to be delivered in order to find paying customers. In other words, creation still is "hard" if you want to succeed.
> I launched something last week. 14 people signed up — no ads, just a couple of posts. 14 real people who didn't have to. That number is tiny and it felt like something. Then I sat down to think about what it would take to grow it and I couldn't look at that math for very long.
This applies to 90+% of founders that have ever launched something. The hard part comes from continuing to push forward when you experience this (which you will over and over). It sounds like the author expects that what was hard suddenly should be easy.
sounds like the author is discovering business.
I didn't read the article, but yes, going from 0-1,1-10 is really hard and really rewarding. And it got easier with the Internet. Going from 10-1k and 1k-1M is a different ball-game. Always was.
The dream of running my own company got me to learn programming. 20 years later I'm an employee at a company still dreaming of running my own company. But now I can realize that reality and dreams are not the same and that's ok. As in, I probably really don't want to run my own company. We'll see =P
In the game of business, quitting = losing.
>> The dream of running my own company got me to learn programming. 20 years later I'm an employee at a company still dreaming of running my own company. But now I can realize that reality and dreams are not the same and that's ok. As in, I probably really don't want to run my own company. We'll see =P
To me having one own's company was just a means to the end: making enough money to live comfortably without the need to get a job ever again for the rest of my life. I too learned programming as a means to achieve that end but eventually realized that I don't need a company if I can short-circuit the path to money. By switching to the right domain - finance, where what I learn might be eventually put to use directly by investing capital into profitable trading strategies.
Back to OP's article, if there's a domain where money as a moat is not a problem, that's definitely finance: https://www.visualcapitalist.com/all-of-the-worlds-money-and...
I work in this domain since almost 20 years and can tell you, noone's gonna risk a billion dollars on crap vibe coded by AI. I wrote before, I don't know what crack these AI people are smoking but when there's real stakes at play, they don't play around with toys. And AI in programming is a toy. The unlikely triumph of "Can I haz teh codez?" CTRL+C / CTRL+V "prompt experts" (mocking it, lol) strategy on Stack Overflow, along with the people who employ it.
I'm not worried about MY particular future in this industry. I'm not worried that AI is gonna replace me, us, or write anything significant here at all in the foreseeable future until it fucking evolves into AGI which is somewhere 5000 years from now, optimistically.
The party's gotta come to an end really soon along with the figures on how much money AI makes versus it's real utility - which is, simply stated, "a toy".
Not necessary but here's my mood while writing this comment: - listening to this song: https://www.youtube.com/watch?v=6EWqTym2cQU&list=RD6EWqTym2c...
It's absolutely wild to see how unevenly distributed the future we're living in is.
That's almost that famous quote by William Gibson: "The future is already here — it's just not very evenly distributed."
Is it any surprise when wealth inequality has been increasing in the USA for the past 50 years? Globally the picture is even more bleak.
Something I've noticed a lot with Twitch streamers and YouTubers is that many of them want the outcome and are prepared to do the work, but want some sort of guarantee of success. It's very difficult to really sell to people that you will work very hard for a while and there's absolutely no promise that you'll ever have more than 14 subscribers. That's simply the core risk of entrepreneurship.
Also, as someone who's main source of income is a YouTube channel: there is a type of threshold effect, where your videos are not good enough to watch until one day they suddenly are.
This means that until you reach that threshold, it feels like you're not making progress, cause every video just gets the same result (no views). Even if below the surface, you're slowly inching closer to that moment where your videos will actually be watched.
I have the impression it's more of a "bait and switch" thing. People get into streaming thinking they can just play games and make money, but then they realize nobody is going to watch someone play game X, they all want to watch game Y, because game Y is popular right now among the people who have the most time to play games and watch streams (kids). So they enter the industry thinking they can just do whatever they want, and quickly realize they have to do infinite things they don't want to actually reach the level of popularity they need.
> Creation has progressively been getting easier since the invention of the computer, it is not a new phenomena. This naturally pushes the boundary on what needs to be delivered in order to find paying customers. In other words, creation still is "hard" if you want to succeed.
Only for developers. Outside of software creation is still hard. Global markets giving access to excellent manufacturing sure does help, but software is a bubble.
Creating marketing material has certainly gotten easier as well, it used to require a lot of work to create these spam pamphlets and company documents but today its trivial. Of course those are worthless to society so didn't help GDP but it filled our society with advertisements and spam and filled companies with worthless documents since now nobody thinks before making one.
The thing is, the barrier isn't near zero. The time to reach an MVP has just decreased. But you still very much need expertise, strategy, etc. to deliver something worthwhile. The bar has just increased.
Seriously you can ship in a week things that FAANGs would have paid billions for 10 years ago.
LLMs are just glue between pieces of your code you still need to be able to plug them into a coherent architecture to do something impressive.
Apple Pay (launched around 12 years ago, and likely costing billions) is exactly the counter-example here. The 'code' was the easy part. The moat was the decade of hardware R&D and the leverage to force banks to adopt a new standard. An LLM might write the API wrapper in seconds, but it can't hallucinate a relationship with Visa and Mastercard. You literally cannot create a new Apple Pay in a week or even years, no matter how much you vibe code.
I'm sometimes baffled by what people think can pass as a product in a real sense.
To the extent that’s true it has much, much more to do with AWS, open source libraries, and collective knowledge, than it does LLMs.
But I honestly can’t think of anything you could do in a week that a company in 2015 would have paid billions for unless it’s something like tweaking an LLM. But in that case it’s the original model, not the 1 week or work you put in.
OK, say you build a Whatsapp clone in a week. How many Whatsapp users will switch to your app?
> things that FAANGs would have paid billions for 10 years ago
such as?
I can only assume the grandparent means Google+
I mean literally anything that leverages modern APIs.
WYSIWG Site Builders, text Chat bots, audio Transcription, voice synthesis.
Yeah building from scratch would take longer, but you can slap a UI, a DB/schema around modern APIs and output something that would be science fiction 10 years ago.
WYSIWYG sure builders existed 10 years ago and did not cost billions. Chat bots were a novelty since the NFT bubble hadn't popped yet, and they would invent NFTs to stake the economy on instead. Audio transcription and synthesis existed and did not cost billions.
You think people would pay billions for a site builder??? Or that it would have been SF 10 years ago? I take it you were not around much 10 years ago.
i would argue that reach has already been the biggest limiting factor for the last 10 even 20 years.
It's just another way of saying "Ideas are worthless, execution is what matters" which has always been largely true.
Yes, you need the idea first of course. But that's truly the easy part. 99% of "ideas" rely on great execution to be worth even looking at - much less paying for - for anyone else.
That isn't what it's saying and I don't think the idea that "execution is what matters" is even true, other than to point out that ideas aren't valuable by themselves.
This is about marketing, about getting people to know and care that the thing you built exists. You can execute perfectly (in terms of making a great product) and not get a single eyeball.
> You can execute perfectly (in terms of making a great product) and not get a single eyeball.
that's a tautological statement - if not a single eyeball is on the product, then you obviously didnt make a great product. after all, who determines a product is great? It's those eyeballs, not the creator.
You've basically chosen to ignore the whole AI argument as if it was just another tool and we had business as usual. Given how pervasive and fast developing it is, there should be an argument why it can be dismissed so easily.
This AI boom is just a hyper-version of previous tech booms (web 1.0, VC, crypto, etc). You have an enormous number of people who just want to get in and build something, but the products they are pumping out don't serve anyone's need or solve anyone's problem.
The moat isn't money for out-marketing your idea that 750 other people are building, it's having a good idea that solves a problem that nobody else is solving well.
I worked during the digital revolution in film, I've told the story a zillion times on HN but basically, I went through the first pure digital film program in Canada, by the time I graduated 70% had dropped out, as far as I know I'm the only one who made a proper go at it, and even then when my startup was taking off, a new hot shot would show up every month and be gone the next when they got bored or frustrated when nobody thought they were special. Tools are tools.
> a new hot shot would show up every month and be gone the next when they got bored or frustrated when nobody thought they were special
I don't know much about film industry, and I have a ton of brainfog from being sick today.
Could you say more? What made them a hotshot? They thought they were like, creative geniuses with digital film or something?
They often WERE creative geniuses with a digital camera! Brilliant people! In fact they were the ones who came and went the fastest, what makes you useful in industry is very frequently nothing to do with genius or creativity or mastery of the tools. It's things like reliability, stability, team spirit, open mindedness etc.
Creative genius is one game, but it's far from the only, or even, main one.
Hope you feel better. <3
yeah this is really a part of it. Both founders and investors get spooked by the rapid entries into a market but persistent not just when it’s hard but when it’s boring goes so much farther.
> the products they are pumping out don't serve anyone's need or solve anyone's problem.
This isn't true though.
Yes, there are too many products being build that don't serve anyone's needs or solve anyone's problems.
However, many of the AI products do solve problems and serve needs.
You're right though, to compare this to other booms, which also had the same problem. This is very much a "hyper" version, which is pretty incredible to be in the middle of.
I don't mean that _all_ AI built stuff is useless, just that the number of products where 'marketing budget' is the bottleneck is dwarfed by the number of tools that aren't that special in the first place.
If you have a product that:
1. Solves a real problem people would pay for
2. Is not trivially replicable by your potential customers or competitors
3. Does not have a natural discovery mechanism by potential customers
Then you need the marketing budget.
That is not most people's problem.
I think item #2 in your list is the real kicker here. Given that AI can write code the threshold for "trivially replicable" is going down.
Unless your thing has strong network effects or a large capex requirement (ex: GPU infra) its easily replicated and I think that's really what makes things hard.
Most business software that was truly trivially replicable with AI, was already trivially replicable with the prototyping tools we had available.
Ok but pick any category of human endeavor and 80% of it is garbage in the beginning. There were 3000 car companies in the 1920s, and most of them sucked, and so they died. The market over time will sort out who survives and who does not.
It will take a few years for investors to figure this out, but in the meantime, everyone is spreading their bets around like peanut butter in order to be in the game.
> However, many of the AI products do solve problems and serve needs.
Every solution to a problem comes with its own costs. It is entirely possible that most solutions that are rooted in modern computing technology have actual or perceived costs that exceed the value of "solving the problem".
The problems that most people have that they really want to solve are not addressable by AI, or computers, or software.
This seems to me like the few booms I’ve seen before. Absolutely crazy valuations with very little behind them, massive hype, everyone’s unemployed uncle suddenly becoming a shallow expert. It’s probably going to end the same way too, once the upward momentum dissipates and things start to retreat to “fundamentals”, we’ll find out that there were a lot fewer solid points in the market than we were all told to expect, so the fundamentals are actually pretty far down. After 5 to 10 years of regrouping, a more mature and solid version will come about and become such a normal part of life we barely even remember what it was like without it.
We are well on our way to the popping of inflated expectations.
Currently people are taking AI hype too seriously and extrapolating its success out in such a way as to discount the value of other businesses.
Example - last week a bunch of trucking stocks crashed 10-20% because a $6M company that pivoted from Karaoke to AI demoed something.
This is just insane. Sure, if say Waymo is pivoting into commercial trucking.. maybe. But people are basically shorting minutemaid lemonade because their neighbors kids opened up a lemonade stand. Demos are easy, products are hard.
> …it's having a good idea that solves a problem that nobody else is solving well.
Added emphasis to the most crucial part, in my opinion.
If you can manage to deliver a product that's meaningfully better than the competition, you still have an edge, so long as you're competent at marketing.
Nothing gets people searching for alternatives as consistently as frustration, and a product that was lazily built with AI (vibe coded or otherwise) is going to be full of bugs and papercuts that make using it a poor experience.
This is particularly true for software that sits in the hot path of peoples' workflows, where thoughtless design, misbehavior, and poor optimization chip away at time the user can't afford to spare.
In short: yes, competition will be plentiful but it will also be almost entirely awful, and capable SWEs can capitalize on that. It won't take much to stand out amongst the mountains of garbage that will be generated in the coming years.
> The moat isn't money for out-marketing your idea that 750 other people are building, it's having a good idea that solves a problem that nobody else is solving well.
This is a very naive take. Any good idea you have can now be cloned trivial in no time at all.
The clones just need to out spend you on marketing even though it is your idea that the LLM cloned.
So where are the AI clones of MS Office, JetBrains IDEs, Whatsapp, Obsidian, etc. so far?
Obsidian has plenty of clones. The others have money as a moat, which is exactly my point.
Which clones? And why not the others?
Which is my point: it seems easy to clone something that seems conceptually simple, but is the result of all kinds of UI, UX, performance, etc optimizations. The reason someone might choose Obsidian over these so called clones is not just marketing, I assure you. The reason people attempt Obsidian clones is that they think creating todo-list management tools, etc is all it takes to implement a PKM, and that that is easy. It is not.
Right now there are kinds of tools I wished existed, that I would pay for, but AI does not automatically provide the insight, good taste, technical excellence, and grit needed to create these products. I could do them, with AI assisting, but do not have the time. It is not a simple matter of saying: Claude, create or clone X, Y, Z.
> Which clones?
Are you seriously contending that apps that have Obsidians functionality don't exist?
Just on HN alone, throughout 2023 - 2025 we were seing like one new TODO app show up on HN weekly!
> And why not the others?
Because money is the moat, and they have it!
That was my entire point - money is the moat.
You have failed to provide a single example of an Obsidian clone.
There are so many of them, aren't there? There's Roam Research (might be the OG one), Logseq (FOSS Obsidian basically), Notion, Emacs' Org-Roam, Anytype, etc. Neovim has like 5 extensions implementing the same idea (such as Neorg), Bram's Vim probably has its own plugins in Vim9script.
You are not making sense.
> Just on HN alone, throughout 2023 - 2025 we were seing like one new TODO app show up on HN weekly!
This response shows you missed my point entirely. I am saying a todo app does not a PKM make! I'm not interested in a vibe-coded todo app, it is useless to me.
> Because money is the moat, and they have it!
You're sidestepping issue and contradicting yourself. I asked: if cloning is so easy, why has no one cloned the JetBrains IDEs, for example?
Remember, I'm not talking about what happens after the cloning. Are you saying no one has cloned Jetbrains because it takes a lot of money to do so? That would contradict your claim that AI makes it easy.
What's the point you are trying to make here? That it is not possible to trivially clone good ideas?
Let's start with my actual claim - "It is now trivially possible for someone to clone your good idea".
I want to clarify your position: Do you think that the bar for cloning someone else's good idea is now:
1. Harder to do with AI,
2. Easier to do with AI,
3. Exactly the same level of difficulty it always was.
Because if you are arguing that #2 is an incorrect answer, there's no real point in continuing the argument, is there? I'm taking #2 as a given, and you appear to be arguing that it is a baseless assumption.
> The moat isn't money for out-marketing your idea that 750 other people are building, it's having a good idea that solves a problem that nobody else is solving well.
An idea is not a moat. Execution is only a moat if being nimble is part of the ongoing offering.
Historically, during booms like this (for example the industrial revolution), it was, because we had patent protections in order to encourage ideas to be brought to market. I don't see how you can have an AI revolution that doesn't just funnel everything to the top without something similar.
Why invent the cotton gin, find investors, and bring it to market if the steel company with the infinite worker machine can instantly compete with you?
how to prevent others from building a copycat using ai?
The discussion here is going sideways, and I blame the underwhelming blog post.
Having money is NOT an economic moat-- i.e., a durable, structural competitive advantage.
He overlooks broader, true definition of moat attributes like labor supply, infrastructure, PP&E, brand, network, natural monopolies, switching costs, regulation. These don't go away with commoditized CRUD apps.
And quoting someone with decades of experience implying that things are hard now and innovation didn't turn over industries in the last 25+ years is a joke.
The more I think about it the more brainrot the article really is. As if all problems are solved and pumping out soulless shovelware companies is worth anything. It really is "just one more app" all over again.
Knowledge ? For b2c it might be more difficult, but in b2b, understanding your customer and their specifics issue and developing something made for them is one of the big challenge. Being able to spit out code for free is useless if you don't know what and who you are making the code for.
The same way you prevented this previously. Copying successful products is nothing new, AI just makes it easier.
Marketing, lawyers, good customer support, creating relationships with customers.
You cant. You can only focus on building your own product and making it durable and just much better
by keeping the how part a secret
Let's just say, building software alone is not enough.
You don't.
You work on niches that have very specific requirements that you can only derive from having a good relationship with customers and so you attend to those needs faster than competitors who are out of the loop.
> you attend to those needs faster than competitors
I wonder if this type of hustling can be called moat building?
20 years ago, people told me ideas were a dime a dozen and implementation was what counted....
It's like that FT chart claiming that the rapid rise in iOS apps is evidence of an AI-fueled productivity boom.
I always ask people, in the past year, how many AI-coded apps have you 1) downloaded 2) paid for?
In addition to that, what they don’t mention is that:
1. Other app stores like Google Play and Steam haven’t seen this rapid rise.
2. There are thousands maybe tens of thousands of apps that are just wrappers calling OpenAI APIs or similar low effort AI apps making up a large percentage of this increase.
3. There are billions of dollars pouring into AI startups and many of them launch an iOS app.
Has steam not seen a rapid rise in AI-asset shovelware?
I'm not talking about the AAA or the AA or even the A space (where AI is being incorporated into dev processes with various degrees of both success and low effort slop), I'm talking about the actual bottom of the barrel.
You never needed AI to make shovelware, you have been able to make a shitty game over a weekend ever since RPG maker was made and there are still games made using that.
AI just helps create some assets for games, it doesn't really make it easier or faster to make games but they might look a bit better.
I can’t speak to the quality of all the games released, but in January 2025 there were 1,413 games released on Steam and in January of this year there were 1,448.
> It's like that FT chart claiming that the rapid rise in iOS apps is evidence of an AI-fueled productivity boom.
I mean, there is evidence for some change. Personally, I'm sceptical of what this will amount to, but prior to EOY 2025, there really wasn't any evidence for an app/service boom, and now there's weak evidence, which is better than none.
> I always ask people, in the past year, how many AI-coded apps have you 1) downloaded 2) paid for?
In the past 5 years, the only "new" app I've added to my phone has been Claude.ai.
Before that I guess DoorDash. And that probably covers the past 7ish years of phone use.
There's just too much shit in the store, a lot of it is scammy or has dark patterns.
For me, "app stores" are largely dead.
Because so much technical functionality has been lost/paywalled/dark patterned/enshitified, I've cut the number of apps I use. I've realized building core personal functionality around the whims of corporations eventually just gets weaponized against me, so I might as well start undoing that on my own terms. Who in 2026 is really bringing in a new app/Saas to do much of anything like we naively did a decade ago? No one I know, we've been shown we will be treated as suckers for doing that.
Well, if someone could create a drop-in replacement for Oracle-DB, Adobe Creative or AutoCad in a month of vibe-coding, I think they could prosper without having their own good idea. I would like to see something like this.
But if you're just talking the fairly simple apps that a single very talented programmer could pump out before this, sure, yeah.
Eh, there's some truth it both. The truth is somewhere in the middle of the spectrum. Distribution absolutely matters, often times even more than the product. And vice versa
No, the moat is having a seemingly good "scalable" idea that solves a problem a few people have and most people think they have. Then getting bought out.
The zip2s and OpenClaws of the world
Buddy, my wife vibe coded an app that solves her PTO tracking problem in 1 shot.
Her payroll company was going to charge her an extra $200 a month.
Go on
Not really. It's money/resources. I had some really useful apps I built for myself. I looked at releasing them, but there were companies whose business model was waiting for people like me to release apps that solve a problem, and then just instantly jump on creating a solution and outcompeting the independent app.
It's trivial for competitors with bigger pockets to outcompete you on your idea, and there are companies whose business model is just that. And with AI customers are trying to do it themselves as well. The only startup I wouldn't be nervous about as a small team without large financial backing would be ones where we start out partnered to multiple companies in the targeted industry so that we can leverage that connection.
Historically this is why we have copyright/patent laws. To make it make sense for people to try to bring their ideas to the world. But with everything changing we are back to everyone just sitting on their concepts/solutions unless they have big money behind them.
In Republic I, Socrates distinguishes the art of medicine from the art of wage-earning. One is about the work; the other is about getting paid. Historically, the craft was the primary goal, and the money was an extrinsic side effect.
Today, the money-making side has staged a hostile takeover.
The attention conundrum is just a symptom of a deeper financialization. Multi-billion dollar companies have turned profit into a data-driven science – analytically turning the screws on every script, product, and interaction to optimize for extraction. This is the destruction of the art of making things.
The real issue is that you cannot compete with an entity that has no respect for the art. When a platform replaces the integrity of the work with the logic of a metric, the independent creator is no longer an underdog – they are functionally excluded. You can be the best at any art, but in a system that prioritizes sheer extraction over excellence, your craft effectively ceases to exist.
It seems that we have forgotten how to distinguish between value and profit, and now celebrate the latter instead of the former. Currency enables ever broader and more niche markets, but the financialization of everything is the Faustian bargain; we gained niche hobbies but lost our souls?
Not only value and profit, but also assets, cash flow and debt.
Altogether in terms of currency, very hard to distinguish in the most meaningful way.
It's too easy for the real non-dollar value element to slip through the cracks, and even end up completely gone, in a system where true value itself is not prominent enough to be recognized as the source of highest growth, like it often happens so many times but has been forgotten.
Success can spoil you too.
Once you've got so much more cash than value at any one time because you've been leveraging a productive opportunity well, and taking profits aside regularly, if you're not careful the currency amount alone can seem like enough on its own.
Uh oh.
I appreciate the conversation the post initiates.
Nevertheless, by counter-example -- OpenClaw's creator was just recruited by people with more capital than countries.
If they could "re-produce" it with their capital, they would've preferred that.
Whatever he has, is still a moat. What that is, is debatable.
Is it brand? Is it his creativity? Is it trust/autheticity? A vision? Ownership of a repo or leadership of that community?
All those are perceived moats (or risks) by these folks that tried to scoop him up.
Luck has always been a solution to reach. It doesn't scale, though.
In the case of OpenClaw I think you're looking at a fairly pure iteration of luck there, too. It isn't even a case of "I prepared for years until luck finally knocked" or any variant like that. It was just luck.
If that is the only counterexample I'd say it doesn't disprove the point, if anything it just strengthens it. Nobody can build a business plan based on "I plan to be as lucky as OpenClaw".
like another guy said, it's rhe marketing thing. the open claw thing became a hit sensation in the news etc. and now OpenAI can claim it as theirs. they have infinite money anyways, so might as well buy stuff like that I guess.
It was 100% the brand. The creativity may be a bonus for OpenAI. They saw a shitload of people buying Mac Minis specifically for this software.
I don't find a single counter-example compelling. I guess as evidence that "only moat" is somewhat hyperbolic?
But to counter the counter-example, what would have happened if he did not join? OpenAI can just write and release their own version. They can then do the typical loss-leader and advertising tricks that OpenClaw cannot.
The "simply write and release" is what used to be a barrier.
He was competition. They can't have people running their own self-hosted LLMs on their own hardware, and realizing it AI be useful without a subscription!
>If they could "re-produce" it with their capital, they would've preferred that.
Maybe not.
OpenAI may have all their people tied up doing more expensive stuff already.
I disagree. I think creativity is still a valid moat. You still need to build good products. Its like a restaurant anyone with some money can open a restaurant but you need to has the creativity to make a good one.
The only moat that can beat the money moat, in fact
Money (better thought of as credit, since we are talking about fiat here) is an attractor so much as it can stand for or purchase productive energy. If that fails (central bank failure, currency failure, government failure), creativity takes its place
You don't need a ton of creativity for a good restaurant if you have good staff, and upkeep. I'll take a boring well maintained and staffed restaurant over an overpriced "creative" restaurant where the waiters are terrible and the chef is even worse.
This assumes the market is not oversaturated. It does not work if there are only a couple of people trying to find a place to eat in a street with 30 similar well maintained and staffed restaurants.
Correct me if I'm wrong but, you may be thinking of restaurants where their defining factor is "having a very creative atmosphere", which will not suit all customers. However it is a differentiating factor which will serve a big audience that is under served in a location filled with "boring well maintained and staffed restaurant".
In my view the creativity comes in in finding solutions to a problem, in a oversaturated market the problem may be "how do I persuade customers to come to my restaurant instead of my competitor?". And following that question may be (in the restaurant example) "what can I offer that is under served in the current market?" The solution to that may be "a biker cafe" or "an overpriced "creative" restaurant where the waiters are terrible and the chef is even worse" (perhaps even rude on purpose).
Additionally I assume you want the restaurant to grow. If you want the restaurant to just survive the bar is lower and you may be able to do that by doing the same everyone else is doing as long as you meet the minimum.
But the argument is that now you have AI for chef, staff and upkeep so its only the recipe writer left. I don't agree with that argument but that is the dream these corporations are selling.
And you'll still probably get beat by a restaurant in a better location or one with better marketing
I can't remember the last time I ate anywhere because of an ad. I eat out a lot too, used to uber eats nearly every single day. I wont reorder at places that mess up my order, but consistent quality I reward. Chick Fil A gives me the best service out of any fast-food restaurant, so they earn my repeat business.
People almost never admit it when they respond to an ad. Even when they very clearly do.
Don't underestimate the effectiveness of advertising, or its ability to influence you.
You don't even see ads for local restaurants, you just go there because its close or a friend recommended it. How could you get influenced by something that doesn't exist? Their marketing is just the storefront, people go there since they are interested in this new place.
They definitely spend on ads but I worked at Red Lobster a few years ago, I definitely remember TV ads. Now and then I get a food ad for a place I'll never go to and never do. Uber Eats is trickier because I just look at the map for whats closest to me.
Google Ads, especially Google Maps, has tons of paid ads for local restaurants. So does Uber Eats and similar ilk.
I particularly despise these ads in Maps because the ad often obscures the search result I'm looking for -- and I end up accidentally clicking on an advertisement for some other restaurant than the one I was looking up.
I distinctively remember when ads have worked on me to buy something. Sometimes I blacklist a brand if an ad is deceptive and makes me click on something, but I don't watch TV much if at all, and the streaming services I do use I pay to have ads removed.
You remember when you remember an ad works on you, of course that is a tautology, you don't remember when you don't realize an ad worked on you but did.
Creativity is always a valid moat by definition. If what used to be creative isn't anymore you need creativity by definition to differentiate yourself from that.
Creativity is a kind of labor. Like all labor, it never earns pay more than once.
Then what are acting and music royalties?
They're the exceptions that prove the rule. For every Taylor Swift, there are probably tens of thousands of songwriters who have never earned a dollar from their creations.
The vast majority of valuable copyrighted/patented creations are works for hire, which is compensated just like other labor. Or the creator signs or licenses the rights away to a publisher and maybe receives contractual royalties, rarely more than the initial advance.
(An alternative answer is "the pay for your creativity is a capital asset. That asset can earn royalties. Usually it doesn't.")
Stock photography?
Font/Typography sales?
What about Pluralsight/Udemy courses?
Substack/Patreon membership?
Buying printable art from etsy?
Yes, all addressed by earlier discussion in this thread.
Your creativity will be copied within days
Some peoples' yes, others no.
What's to stop someone copying your creations? Creativity is the reason a moat is needed.
I think creativity is still a valid moat. And I think creativity will become a very important moat!
I am a firm believer of "limitation breeds creativity" and being able to make what you want with immediate result will lead to less diverse solutions. In a world of in-diverse products you need more creativity to stand out as it will become harder to conceive of something "outside of the box".
Similarly to how anyone can basically create any image they desire in Photoshop (with some limited training). Leading to a lot of images of a similar style instead of a lot of different styles (ai slop anyone?). This is because between the idea and the result the roadblocks are reduced, the process is smooth (tools aim for 1:1 conversion). In the creative process these roadblocks are usually where you will find interesting new directions or ideas. And very often the original idea was not that interesting to begin with (and perhaps not as original as we would like to believe).
The claude.ai web UI has a bug right now. If you inline some code by typing an opening backtick then the closing backtick swallows your space, puts the next letter you type after the code, then everything else back inside the code span.
One day we might be able to write software without bugs. That day is clearly not here yet.
(Firefox in Linux if anyone wants to repro. Can’t file a bug as it’s a closed, proprietary piece of software.)
That's my hope, too, since I've had the unfortunate luck of practically being forced to fulfill my lifelong dream and create a software company as a solo dev in 2025. Great timing...
However, I've gotten doubts about the value of creativity and having a vision. Plenty of people have great creativity and a fantastic vision, and we can see already how AI allows people to churn out interesting and useful projects at much faster pace than just a few years ago. The problem is that it seems more and more impossible to actually make a living of that.
Anybody with the money to have access to enough AI will be able to create a clone of software X. We might not be there quite yet but it's just a matter of time. That seems to be a death sentence for any small software business.
I think, in time, it will be shown to be the real, persistent differentiator. And far more valuable.
But it's also not a moat in the same way. It's accessible to everyone, but you have to actually disregard the parts of yourself that want to drive hard in some direction just for money or power or external validation.
From the look of things right now, it may take some pain before that really gets to shine.
This article sounds like "I can't make a profitable SaaS, hence others can't too, and those who can, are using absurd amount of money or prior reputation to do so"
I advice to grow up and do some proper research on what problem to solve, and how to build trust with audience before pushing your products.
I dont have much money or prior reputation when starting out, I blogged weekly, then slowly launched technical ebooks, one time utility app, and finally SaaS , took me about 6 years of showing up every day to have a job-replacing income.
Your other post mentioned you gave up and declared failure after 3 weeks of launching your SaaS, I think the timeframe expectation you had might be unrealistic.
Doing hard things has always been, and always will be, hard.
Building a static HTML page was “hard” in the 90’s. It took actual skills.
Any piece that gets easier automatically opens up more hard avenues to tackle.
no one is willing to pay you for easy.
I was there in the 90s. I built a few bad static HTML pages. It wasn’t hard. There are lots of stories of non-CS / non-technical people making stuff from the dotcom era.
Making a dynamic page was harder. Integrating with a payment system was almost magical; there’s a reason PayPal became big.
But what was truly hard, and continues to be hard, is building a page, either static or dynamic, that people actually want to visit.
Building static pages that worked in both Netscape and IE 4, and could function well with constrained dial up speeds, may not have been "hard", but it did come with a number of challenges.
Ahhh that lines up with a thought I had recently: you get out of life what you put into it. I am beginning to believe that there's some kind of metaphysical rule here that is true everywhere, all the time.
So maybe the solution is: find the hardest stuff to do and do the crap out of it.
I really don't think that's true, you can put a lot of effort into a wrongheaded strategy, netting you no or bad results.
Conversely, a good strategy can get you good results without that much effort.
Wrongheaded strategies that net you bad results, often result in lessons and ideas that can be further pursued on their own merits.
The followup thought / concern that occurs: what if the number of hard things is going down?
I would have imagined it's going up while the percentage of people who are capable of doing the hardest things is going down.
Not that I know much of anything.
We probably need to split apart things like "hard" and "both hard and useful".
Just because you're doing something hard, doesn't mean anyone wants it.
Just because you're doing something useful doesn't mean you're going to get paid much for it.
Just because something is hard and useful doesn't mean someone is going to pay you for the cost of the effort.
The entire pitch of LLM companies is that that’s not true. The LLM does the hard work, and you pay the LLM company for the tokens. So the gate is just, can you afford enough tokens.
Not saying that’s what will happen in reality, but that is the marketing pitch
Coding agents are good, but once the complexity is high they're not good enough. Eventually your agent won't be able to make changes to your code base without introducing bugs with every change. In my experience, the agents aren't very good with abstractions yet, and no amount of testing can completely paper over that problem. So yes, the industry is changing dramatically and at a breathtaking rate, but I don't think money is the only moat left.
Not just complexity, but also anything requiring any actual ingenuity. It's impressive how "junior" it can be, with its (current) statistical shackles. Worse, if you flat tell it what to do, if your approach is too far into the statistical weeds, it'll flounder like crazy.
what is an example?
Just tonight I was trying to get Claude Opus 4.6 to design a lock-less data structure, and it kept failing to solve problems which I kept then solving for it, but once presented with the solutions it was able to check them, and I think correctly so (that's not nothing!).
There is an old word for it: saturation.
And let's be honest that's not a new thing. It's been already a long time since you had a revolutionary idea in the shower only to google it(or use an LLM nowadays) and discover that there are already eight different apps that do what you were thinking.
The way to tell if a business has a moat that I once learned was, "If someone gave you a billion dollars, could you go compete with that business and have a reasonable chance of winning." If the answer is yes, then the business has no moat. The numbers are bigger now, but I think the principle remains the same: money cannot be a moat.
A moat is around something that exists, meant to protect it. You're describing something else, not a moat with your example. OP used moat correctly, creative effort around existing earned skill, brand, etc.
To answer your question. Yes to a player already in a market with lack of funding, a billion dolars could be the necessary moat to win.
Money was, is and likely will be a moat for a while. But as a proxy, it may not be enough as a moat. Scarce resources may require more than money— e.g. IP classes or, if you're China, ASML machines.
I can't think of a single software company that has a moat then. Good rubric to have and makes sense.
These types of takes are getting really tiring. The game isn't over, just the metagame.
I'm not saying I have the next metagame figured out, but AI doesn't magically solve all the worlds problems by existing. It likely even creates a new classes of problems. In such an environment there's going to be opportunity.
Are the software shops and devs who were winning at churning out SaaS products going to be the best suited to this new environment? Maybe not, but adaptation wins over non-adaptation, something that has always been the case.
The well is drying. You have less money available for hustling and less small players with money. With layoffs happening everywhere you have more people with ambition and time on their hands, but less of them can afford big expenses or major risks.
The hype machine currently pushing for agents is selling agents ability to do automated marketing. However the bigger companies know better than to create giant security holes and the small players are either not technically skilled, or will balk at the huge per-use fees for the good models, or will be drowned out because of low quality cheap model output.
I had this conspiratorial thought, would a model really just spit out some money-making scheme, or would it be blocked. Made me think to run the model myself on my own GPUs but still a black box, at least you control your prompts/data flow locally.
The OP should read up on business strategy, probably starting with Porter's Five Forces.
Hint: Money is a nice thing to have but it is very definitely not a moat.
If a company is making above market returns and the only thing stopping a potential competitor from competing with them (aka the company's so-called "moat") is that "it takes money", that company does NOT have a moat.
It's very easy for a potential competitor to calculate that, after x-months or y-years, they will have made enough profits to pay for the cost of building the competing product. As long as that amount of time is finite, there is excess profit for a competitor to take, and the company will find it's so called "moat" wasn't a moat at all.
This isn't a new thing. It's been a fact for centuries or millenia. It's one of the many things that makes success in business hard.
Porter's Five Forces is one distillation of the foundational principles on which moats can be built (and yes, this is a non-trivial subject, so success in this area generally does take more effort than just reading or skimming the Wikipedia page, but if you had to distill it down to one sentence, it's probably "try to build something that has network effects").
> Hint: Money is a nice thing to have but it is very definitely not a moat.
in this context i think where money becomes a moat is when you consider Google can keep advancing their LLMs and tools while operating at a loss because they have the cash and revenue to do that. They'll just keep going and wait for everyone else to go bankrupt and then the whole AI market is theirs. In that way money, and the ability to out last your competitors, is the moat.
This person needs to study Ronald Coase's Theory of the Firm in order to not make incorrect assertions like 'the only moat left is money'. There's many other criticisms to level at this piece, but this is the one I'll use my reply/comment on. If the author was right, then the entities with the most money, or ability to print money, a.k.a. the government, would be in charge of all economic activity. They are not. Empirically, what we observe about human economic activity is that it's not about who has the most money. Money helps, and certainly there's economic structures that are a tier below government but definitely far larger than a small business that can use money as a weapon to stay entrenched or gain other advantages in the market, etc, but it's not the only thing. Empirically it is just not. There has been, is, and will continue to be room for what other commenters have called 'creativity' broadly speaking, and that even includes creatively recruiting other people to your cause/company versus another's, etc, that 'the one with the most money' has not alone shown to be decisive.
> The people winning mostly had a head start. Or they have money. Usually both.
It feels like that doesn't it? But, as one counter-point, OpenClaw. :)
Btw I did a deep-dive into AI moats last week and wrote a blog post about it. Relationships were most likely the strongest moat from my research - but definitely having a large amount of money in reserves helps. https://www.cjroth.com/blog/2026-02-11-moats-in-the-age-of-a...
This is not new. When I started my first company in 2012 it was bootstrapped and getting anyone to pay attention to what I was building was almost impossible. I had to pound the pavement and meet people in person at coffee shops and pitch to get my first few users.
Then when I raised from a16z and had some money in the bank it didn't get any easier. The money didn't help (maybe it wasn't enough). Ad spend or content marketing or paid channels were all hard regardless of the free vs paid.
Maybe I just wasn't good at it.
That was before AI and you had to manually pound the bits into place.
Now with AI yes there are a lot of people shipping a lot of things but humans can tell when someone's put effort into something vs not and the time to traction is still as high as it always was.
Someone should do some analysis on number of things that go "viral" or gain adoption quickly today vs 5 / 10 years ago.
Getting traction has always been hard. Thats just business.
This is certainly one conclusion that could be drawn.
Another conclusion could be that as building software gets easier (like it did for ex in the 90s and again in the 2010s), opportunities are created for new entrants to displace Bad Old Software.
Those expensive Enterprise apps that everybody hates? Are absolutely begging to be replaced by something better for half the money.
We still live in a world where most individuals own more compute power than most universities did in the '80s, yet the only sign of automation is useless push notifications.
Data behind one pane of glass can't easily be moved to data behind a second pane of glass. Simple stuff like "move my Instacart shopping cart to Costco.com same-day" is a manual affair. This is a subset of the general problem that more apps has resulted in more data silos that are generally isolated, without APIs, without automation.
There are zillions of problems out there for which people will pay money, but money chases the same 4-5 problems at a time. Just work on one of the other ones.
I've been enjoying smashing through those panes of glass with LLMs. Just spin up a browser, go and export all the requests to undocumented user endpoints to curl commands, specify what you want out of it and after a few minutes you have an api endpoint up and running!
> as building software gets easier [...] new entrants to displace Bad Old Software
This didn't happen for music.
It is much easier to create/record music today than in the 70s and 80s, but the music created today is mostly boring AI music and not new exiting/inventing music.
Try finding some small artists you like on soundcloud, see who they follow and who follows them, I'm sure you'll find something interesting
> the music created today is mostly boring
One thing hasn't changed: people get older and hate on music that the youth like.
> The value of human thinking is going down
The value of human repeating other human's know-how through hard work and thinking is going down. Unfortunately (or fortunately?), many types of programming is one human repeating other human's know-how.
> The value of human thinking is going down.
Wrong. Creativity, innovation, intuition, taste - all forms of thought solely inherent to humans, all going up.
I used to agree completely, and a part of me still does. But the part of me that's scared that this isn't true is getting louder and louder.
I'm a novelist and software engineer. The value of one of those skills is trending to zero. I'm not seeing much to suggest, in the face of the hockey stick of doom, that the other isn't.
If anyone can create what they want won't coming up with good ideas be the value, therefore the creativity will become more important?
I don't know what you wrote and I agree that it is very sad that the skill of writing is very undervalued at this moment. But I think that trend is not new and in reality not created by AI (just accelerated). But I think value of writing was never in the writing itself but the knowledge and ideas behind the writing. Because mechanics of writing sped up (handwritten, typewriter, word processor). But the process of writing a novel did not speed up significantly as the bottleneck is the ideas and iteration on those ideas not the delivery of the output. Similar to how programming is more then typing code.
So I think the real crisis right now is that people equate typing speed with programming speed (or writing speed) and due to advances in the former undervalue the later.
Sadly the reality is that it won't pay for the bills anymore.
Perhaps in a while people will realize how important the quality of ideas (and the iteration to arrive at them) is.
Depends. Are you still measuring value in dollars?
I got mouths to feed, so... yeah.
Indeed. But the line of discussion was about the value of human thought.
If our human thoughts can't be translated to food and shelter, then we'll pick up guns and go steal the food and shelter from someone else.
Noted.
Sometimes what needs translated to food and shelter isn't your thoughts, however. If you nurture and value thoughts regardless, this point will be understandable.
> Wrong. Creativity, innovation, intuition, taste - all forms of thought solely inherent to humans, all going up.
How do you define value?
I mean, if the demand remains the same and the supply goes stratospheric, the value has to go down.
After all, purchasers don't very well care that your product has creativity, etc and that that other identical product was just an AI cloning your product.
why? We have no real evidence to support this claim.
> a paid, invite-only social network where every person is verified human and there's no algorithm
This seems like an incredibly niche product that only a handful of people are interested in to begin with. It isn't an notable or surprising result that building it resulted in little interest from general audiences.
Kind of ironic - the moat is money...
At the same time, I see the appeal. I feel like 10% of the comments I read lately are "is this an AI response?" - would be nice to be free of that. Probably not possible tho.
It also strikes me as being in competition with, you know, a group chat.
i was just about to say, what they're describing is an imessage group chat with your friends.
I think the other moat is access to non-public data. If you can train, measure, or make decisions based on specific data that the vibecoder trying to clone you can't get, you can keep ahead.
One thing I’ve noticed doing Show HNs recently:
People are flooded by new projects and assume (rightly) that most are low-signal, so they don’t engage. Because there’s low engagement, new projects get even less visibility. That reinforces the belief that nothing interesting can be built anymore.
I ship earlier now (often free and open source) to learn faster, but it doesn’t change the attention dynamics much.
The bottleneck isn’t building. it’s distribution and who already has an audience. Now dont get me started on getting an audience. thats a whole different pain.
I commented this earlier and it was instantly downvoted for some reason, but I totally agree with both points; I also think they are intrinsically connected in many ways because using customer-development to understand and build an initial audience often helps identify distribution channels. Here are a few examples I have found
1) I made a product to help people manage their diets; crowded space so it was hard to find users, following Steve Blank's customer-development process I found that the product should have been made for Dietitians not for end users. This may have been obvious to most but it was but to me at the time.
2) I made a product for Medicare/Medicaid but it was hard to get my foot in the door to even be considered for any of the Gov. contracts, so following the same model again I found that if I used politicians that had made promises during their campaign that my product could help them to fulfill then I could use them as distribution channels.
One thing I disagree with or don't understand is
> ship earlier now (often free and open source) to learn faster, but it doesn’t change the attention dynamics much.
I have found the opposite, it seems like the bar is much higher now and not even the boost in productivity that I have gotten from using AI has allowed me to ship secure/high quality products as fast as I would like to; perhaps I am just too insecure about my work but its all new I guess.
Interesting article! FYI I clicked on joinkith.com and got: "Application error: a server-side exception has occurred while loading joinkith.com (see the server logs for more information)."
If someone has a problem they need solved:
* Attention span won't stop them recognizing a solution
* Numerous solutions won't stop them adopting one
* That the developer put little effort into building that solution won't put them off.
The real answer is of course that there's a lot of stuff that being built that doesn't solve people's problems people have, either because it targets a problem that doesn't really exist or because it fails to solve the problem it does target.
Of course people will buy software.. but the 80% margins we had of yesteryear won't be there
The result of the barrier to entry being erased is that prices will also be driven to 0 as products are commoditized
It's hardly worth going into a market when you're 1 of 1000 and profit margins are at 3%
Any creativity you add, new spin you put on it, new features, innovation.. all that has negligible cost to copy
I don't think we're quite there yet, but this is where it's quickly heading
Money is not a moat since you can buy money (debt) or sell equity for money.
>The thing I launched last week is called Kith — a paid, invite-only social network
Social networks already is an existing competitive space and making it both paid and invite only obviously will hurt its adoption. I wouldn't have been surprised if this failed to get users even if it was free and even if this was preLLM. A brand new social network doesn't truly solve users problems.
The only real mote was and is mindshare. Build mindshare, don't screw your users, and your competition will have a hard time dethroning you. LLMs will probably force SaaS and software prices to come down, which will be a great opportunity for dethroning reigning champions.
PostgreSQL, for example, has a ton of mindshare. It will be hard to dethrone it.
OpenSolaris was Sun's attempt to recover from the loss of mindshare to Linux. The a company that had become dominant in databases due to mindshare acquired Sun then failed to understand the mindshare play, and it killed OpenSolaris. Explain that one to me.
Even mindshare might not be such a moat now.
AI coding assistants make it trivial to switch between models, and developers will switch to a different model from a different provider if the new model is perceived to be better than the current one even if that perception is completely subjective.
Furthermore, the only users who will actually pay for AI are these assistant-using developers and other power users -- people who will not hesitate to switch to something better. We are already used to new model releases being neck in neck and there is no real incumbent/underdog dynamic going on here.
Yes, there is strong mindshare among the normies, who only know about "ChatGPT", but as soon as they graduate into a powe user, then the above applies to them.
I liked the part about attention being the scarce resource now. Everyone is competing for your attention. But then I see a world in which openclaw is managing emails for people and searching the internet for them and shopping in their behalf. How long until we start seeing advertising specifically targeting AI instead of humans?
Not discrediting the post, which I think is worthwhile and generally pointing at the right thing, however:
> I launched something last week. 14 people signed up — no ads, just a couple of posts. 14 real people who didn't have to. That number is tiny and it felt like something. Then I sat down to think about what it would take to grow it and I couldn't look at that math for very long.
They launched a paid social network, with no content available without joining a waitlist.
This would not have worked 20 years ago either. Bootstrapping the content for a _free_ social network is incredibly hard. But a paid social network where the only differentiating factor is that users are humans, and there is no activity in the network? Not going to work.
>as someone who's been building for the internet for 25+ years, this is the first time that i've ever felt like it's very difficult to make money building new things.
I understand it was already the trend 25 years ago, but way before that you really weren't expected to be able to make money building things for the internet.
The internet itself was simply not designed for that to begin with.
Building things where the internet was an element was already getting bad enough.
The force from within to return to "normal" baseline may yield, but probably never go away.
>The people winning mostly had a head start. Or they have money. Usually both.
As said every millennium since institutions and finance have existed.
>Show HN, the one place the internet was supposed to notice if you built something real.
No no no no no. This is for people who want to share with a much more limited audience than the entire internet.
HN readers did notice a lot of times especially when the project is amazing, OTOH sometimes the latest little side project from somebody well-known, or random interest could be shown.
Naturally the most popular things are free since that's inherently the most compatible with the internet anyway.
But real marketing and promotion is supposed to be far away from this site. If you're trying to sell to "the internet" you've got the whole rest of the internet for that.
HN is not supposed to be enough to be widely noticed at all, if you've got something that's worth marketing, YC is there the whole time and might be able to get you making the most of the internet and then some. Especially if you need a moat of money.
But why do so many people think the only business plan is to prepare to be sieged by a small enough horde which can be deterred by a moat anyway?
>if you're not already moving, you might never take off.
>The cost of acting like it isn't true when it is: permanent.
As I first mentioned, the internet being in place so people can make money off of it is the thing that just wasn't true to begin with, lots of people had some pretty good workarounds for a while though.
I've been watching businesses from startups to large corporations lock in high costs the exact same way for decades before the internet ever came around.
For the comment from "Is Show HN dead? No, but it's drowning", the article author editorializes
"One of the great benefits of AI tools, is they allow anyone to build stuff... even if they have no ideas or knowledge.
One of the great drawbacks of AI tools, is they allow anyone to build stuff... even if they have no ideas or knowledge."
into
"One of the great benefits of AI tools is they allow anyone to build stuff, even if they have no ideas or knowledge. One of the great drawbacks is they allow anyone to build stuff."
which removes the rhetorical effectiveness of the comment (and also breaks the promise of a quotation). I recommend that OP represents the source exactly.
____
I now see that this article contains multiple GPT-isms
Oh good catch, fixed.
It would be better if the article editorialized it thusly:
> One of the great benefits as well as one of the great drawbacks of AI tools, is they allow anyone to build stuff... even if they have no ideas or knowledge.
I'd paraphrase it with "said that" since the quotations indeed present as though verbatim.
I disagree. Customer relationships, taste, creativity, tenacity, execution excellence, industry relevance, reputation, non-public data, etc... are all hard earned or intrinsic capabilities that matter a lot more than feature development speed/cost.
I thought your blog sofware was interesting, and you have a blog on it from last year. The Git link on this page: https://elliotbonneville.com/building-a-blog-in-90-minutes/ however, is broken.
Broken Git link: https://github.com/elliotbonneville/elliotbonneville.com
I really like this article, particularly the part about the gravitational threshold. My overall thoughts about launching products is that there will just be more products with smaller audiences, similar to how streaming broke down linear cable. As a software entrepreneur, this means going wider not deeper. Use your human skills to build relationships with people who have that gravitational pull. If you don’t have those relationships, go create them IRL. For context, I’ve founded and sold a social network to AOL, helped raise over $100m, and had 2 other acquisitions.
Um... can we talk? :D
Utilizing connections with influencers seems to be the most effective way to break through the noise now. Ironically, that's how it used to be before marketing and product development became so accessible.
It’s been true for some time. That moat has been getting shallower and shallower everywhere it isn’t inherited wealth. Not entirely shocking, large amounts of human history have been organised that way.
This post was an Ad also for his paid social network, it did reach me because the message resonates with me as a builder but I don't want to sign up to a social network
So the attention was there but not the conversion.
> Creation used to be the scarce thing, the filter. Now attention is. Most of us are on the wrong side of that trade.
"Wrong side" means that people on that side have to work harder to reach us. The terms of the trade did not change because creators are not starting to pay us to get our attention. Well, it happens sometimes: free trials, free tiers, coupons, etc. but it's a well established practice. The post contains a reference to "do more marketing".
"the value of a human eyeball" / attention is and always will be the limited resource. But I wish the way the economy worked wasn't that attention is sold for money, which makes money the moat, and sets a floor on how low-priced things can get for customers too. Is this really the best the economy can do? Or is it possible to have a fair LLM-based search engine that matches customer need description with stated product descriptions from providers (while weighing customer reviews, etc)?
Kith looks interesting, but the main problem there is verifiability. How would it keep bad actors out? How does it verify posts by invitees aren't AI generated?
It does sound interesting and I signed up for the wait list. But I actually don't like pure chronological order. It feels like I have to look at everything to find the good stuff. Here I probably won't miss out too much if I check the front page once a day.
I hope this is true, I'm trying to make a multi-agent orchestration thing that looks at grain futures, satellite imagery, news, etc... to trade crypto. I'll probably lose but yeah.
Maybe the guy doing their 9-5 can run many agents to make them money while they work their day job.
Is that a thing, you get hired at some company then you use an agent to work for you, deep fake video calls, cursor code... that would be crazy. Get another job and split your time between agents for minor corrections.
>Every morning a few thousand people wake up and ship something. A tool, a SaaS, a newsletter, an app that does the thing the other app does but slightly differently. They post it on Hacker News. Nobody clicks.
>This is not new. What's new is the scale. An AI can wake up (or whatever it does at 3am) and ship twelve of these before breakfast.
That's fun, I'm sure if somebody actually checked that and graphed it, you would not be able to pinpoint when AI starts on the graph
Here is the Show HN Graph. I think it is obvious when AI starts. https://petegoldsmith.com/2026/01/26/2026-01-26-show-hn-tren...
Ironically, lots of people are going to get pushed back into agriculture because yields are going to fall and humans can pick marginal/mixed/non-row crops much more efficiently than robots, plus local consumption eliminates logistics costs.
So yeah, techno-feudalism is going to drive us back to agrarian feudalism.
And the super rich now even switch to other people's money. A fund makes investments in SpaceX and Anthropic available:
https://finance.yahoo.com/news/retail-investors-access-space...
It would be hilarious if the final "IPOs" will be in SPAC form with the help of SPAC king Chamath.
I partly agree. Yes, the entry level is low. It helps me to gain time on my life, doing other things. But I do not believe that much that more people are going in. Most non coder people i know will not code a thing. Most bad student i got will not start to do side projects. So, yes people wanting to ship something ship more, but for the rest, there is still place.
> The corollary is rarely mentioned: the value of a human eyeball is going up, because there are only so many of them and there are now infinite things that want to be looked at.
There's an easy fix for this. Agent to Agent SAAS
> The value of human thinking is going down. You probably knew this. The corollary is rarely mentioned: the value of a human eyeball is going up, because there are only so many of them and there are now infinite things that want to be looked at.
Hey, so I'm thinking about getting my car washed..
This article reads as overly hyperbolic; cashing in on the AI hysteria. AI derangement.
I am not cashing in on AI hysteria. I am AI hysterical. lol. Kinda joking but kinda not. How do you have chill while the world is crashing down?
I’m young so that’s probably part of it, only ~12 years into my career and haven’t experienced too many world defining shifts.
The world is crashing down for a huge variety of reasons, none of which are LLMs stealing jobs or anything like that.
I like the idea of you https://joinkith.com/ but I don't see how it can possibly work. These are the issues:
There are people who are sick of social media and will never be convinced to join up again. They've already left the building and aren't looking for anything else. I'm not quite there, but almost.
Other people are using established social media simply because that's where their people and orgs publish. I am eternally frustrated when my local cafe uses Insta or FB for their "web presence", but I'm not going to be the one to convince them to use something else. I hate that my local rock climbing partner finder group is located on FB, but what can I do about it? I also think it sucks that there are thousands of people in that group - I soon realized that this group simply doesn't work for me, since rock climbing requires high trust and I can't trust thousands of people.
Many people resist the idea of signing up for yet another social media account, esp when none of their people/orgs are already there. For example, I've sometimes thought of starting a Heylo group for local rock climbers to find partners - this might actually help me find more climbing partners. But I've never tried it. I just don't think people will join. The barrier to entry is (1) install app (2) create login (3) use app. SFAICT no one wants to do this if they're already on FB and already are a member of the group there. Even people that I know manage finding partners with email lists (gag). Can you imagine how much higher the barrier to entry would be if adding (4) you have to pay a monthly fee?
I do like the idea of "only allowed to invite someone that you know in meatspace" but how is this enforced? I also recognize that requiring payment could help increase the trust level, and I recognize that members have to pay in some way (ads, fees, sponsors, privacy violations) in order to support the platform.
> I’m young so that’s probably part of it, only ~12 years into my career and haven’t experienced too many world defining shifts.
I'm old (52) and a bit AI hysterical despite being well aware of the reasons I supposedly shouldn't be (variations of Jevon's Paradox, the fact that we've had similar disruptions before, etc).
I can't help but think that both the speed and massive breadth of the AI disruption across so many industries all at once makes this a very different risk than anything we've experienced before, in my lifetime or before it.
It also doesn't help that at least here in the US this is all occurring when our government is both openly corrupt and particularly dysfunctional at solving any of the real-world problems facing its own citizens.
looks like the real tension here is that "money" and "reach" aren't quite the same moat though. I think the post kind of conflates them. existing audience is the actual barrier - money can buy ads but it can't buy trust or distribution, not quickly anyway
the gravitational threshold thing is real ngl. I've seen the same dynamic in product launches - identical quality, completely different outcomes based on whether you're already above the line or not. that part holds up
not sure the "creativity is the moat" counterargument fully lands either. yeah taste matters, but AstroBen's point is valid - anything that gets traction gets cloned basically immediately now. so creativity gets you first mover advantage for like... a week?
maybe the actual moat is just community? people who already trust you before you ship. which is a form of reach I guess, so kind of proves the point
So where are the AI clones of MS Office, JetBrains IDEs, Whatsapp, Obsidian, Sublime Text, Claude Code, etc. so far?
I think there's still value in building quality products, but AI makes it easy to build something that appears good but doesn't actually work that well. It's very difficult to communicate the thought and intentionality that went into a well-designed product in a way that stands out amongst the noise.
Why are HN mods appending a question mark to the original title which is obviously a headline of a blog post, i.e. a thesis of the author, and not an assertion of fact? You may as well suffix every headline with a question mark.
http is now a trash can no by AIs activity, they only reveal to our eyes something that has been already happening for years, not perpetrated by robots, but something worst: marketing industries. But in other hand, why to pay for another social network when exist technologies like gopher, RSS feeds or gemini-protocol?
Let me guess, there's a link to his soundcloud or something else he's pitching at the bottom?
> the value of a human eyeball is going up
There must also be some value in advertising to the "eyeballs" of the AI? Even if they don't (yet) make that much decisions about spending, they do influence human decisions.
It's the new SEO. Enterprise software vendors are already tailoring their marketing pages and docs to influence AI for when some VP asks their chatbot "which XYZ software should we consider buying?" This is something a coworker is facing at work as we try to leverage AI to speed up market discovery/tool selection work.
There should be more tax on capital because they are using it against us.
If you thought things were hard now just wait for the industrial-scale fully automatic fast-follow bots that will nearly-universally nuke the human-created original product to oblivion in a few years...
I saw Peter Steinberger whose creativity is huge and made a difference. Yeah you can say he's already rich.
But I also saw many people like him including the author of Flask. Also the author of XcodeBuildMcp, tailwindcss
Not sure I agree with this - if productivity is going up, then it seems like the value of thinking would _also_ go up
Each thought is just worth so much more
Are you not concerned about getting a copyright strike from the fashion brand?
https://kith.com/
Your concern is valid, but you are misguided on a few details. I will inform you as best I can in the context of US intellectual property law.
I am not a lawyer, but I have spent a lot of time adjacent to this stuff, and am not aware of a "copyright strike" as a legal concept. I assume you are borrowing this terminology from various platforms who have their own internal "three strikes" policy for user-created content that violates copyright. However, those policies are entirely for-and-within those ecosystems, and do not directly correlate to any legal notion of "copyright strike". If you violate the rights of someone's work, they send you a C&D and/or sue you, regardless of the platform's use of some "strike" system.
That "someone's work" is key: Copyright covers the output of creative works: writing, art, photography, code, music, video production, et cetera. Copyright law would apply if social media Kith stole an image from fashion Kith's website. It would also apply if one was sent social media Kith the source code from fashion Kith's website, and social media Kith used it (I don't know why they would, but it would violate the copy rights of fashion Kith). The violation would be about the work, not the name.
You referred to the brand name, which would be covered by Trademark law. Trademark law concerns itself with more than just the name, but also the design/logo, sound of the name when spoken, etc. Within that, there is a focus on the likelihood that the public would might confused and the degree to which one brand or another might be damaged by that confusion.
This leads to (somewhat common) situations where there are two marks with the same name, but in different industries, because there is no risk of brand confusion. For example "Delta" is a trademark of an airline company, and "Delta" is also a trademark of a faucet company. They have different logos and products/services in different industries (the USPTO has 40-or-so categories), thus there is little risk of brand confusion.
On the other hand, if two businesses have products/services in the same category, even a different name can be a Trademark violation if it (for example) sounds similar enough. Imagine a hypothetical hard drive manufacturer named "StoreTech". If you started a hard drive manufacturing company and named it "StorTek" it's very likely that StoreTech would pursue trademark action against you, even though the name is technically different!
There is A LOT more to Trademark than just this, but these are some factors to keep in mind.
This whole thread is classic HN. Someone uses a random example to illustrate a point, and the rest of the post is about that.
The definition of a moat is what cannot be bought.
Everyone is excited (or scared) by vibe coded startups, but ask yourself, how many vibecoded apps do you use weekly ? Every single one I've tested looks great at first, but is an UX nightmare with awful performance.
Even worse, my founder friends are all churning slop prototypes with Claude Code with zero product vision and are going nowhere. Is this revolution just a mirage that is FOMO-driven ?
While building has becomes way cheaper (and probably is going to become even cheaper in the future), is building something exceptional really that much cheaper now?
AI has certainly made it so much simpler to just pump "something" out (slop), but did it actually make building something that went through hundreds and thousands of iterations significantly cheaper?
I also like to think AI is really raising the bar for everybody. In the past, you could easily get away launching a product with a crappy landing page and a couple of bugs here and there, is that still the case? Don't people just expect a perfect landing page at this point (when's the last time anybody specifically talked/ thought about responsiveness?) paired with a flawless onboarding etc.?
Building was already not the main obstacle before AI. Distribution was, and remains so - more than ever.
> When creation was hard, skill was the differentiator I don't think people use things because they are hard to make
The author of this article is missing whats really going on:
Hard novel things are still hard and valuable, easy things are now super easy and really aren't valuable anymore.
Lets take an easy thing... like a meal planning app or a todo list or something. Its what we would all call a weekend project, but in 2022 it would be hard to develop, get the legal stuff reviewed, do the marketing, and do all the rest of the operations in a few weeks. That stuff is basically a series of prompts now.
Now lets look at a hard thing, such as decreasing the cost of titanium production (courtesy of orca: https://www.orcasciences.com/articles/there-has-to-be-a-bett...)... to be able to scale operations and bring the cost down you are going to have to do some real science and engineering... LLMs can certainly help you, but this challenge is a far cry from something that an LLM can handle.
*Novel things have just become MUCH MUCH more valuable, slop is way way less valuable*
> When someone suggested the answer was marketing: > jUsT dO mOrE mArKeTiNg!!!!!
This is a good point. If there's a problem reaching people because the information channel is saturated, the solution is to increase the information? And then everyone reaches the same conclusion and increases.
This destroys the channel. It's not a zero sum game. If everyone markets, nobody will make the sale because the customer will nope out and see nothing.
What's wrong with screenshots? Those X widgets are cancer.
"The value of human thinking is going down."
No! I fundamentally reject this.
The value of unoriginal thinking has gone down. Thinking which is quotidian and pedestrian has become even more worthless than it already was.
The value of true, original human thinking has gone up even higher than it ever has been.
Do we think no new companies will ever succeed now? Of course not. Who, then, will succeed? It will be innovators and original thinkers and those with excellent taste.
Why did stripe make big inroads in developer spaces even if they are in an ultra competitive low margin market? They had excellent taste in developer ergonomics. They won big not because they coded well or fast (though I know pc thinks their speed is a big factor, I think he is mostly incorrect on that) but because they had an actual sense of originality and propriety to their approach! And it resonated.
So many other products are similar. You can massively disrupt a space simply by having an original angle on it that nobody else has had. Look at video games! Perhaps the best example of this is how utterly horribly AAA games have been doing, while indie hits produce instantly timeless entries.
And soon this will be the ONLY thing that still differentiates. Artistic propriety, originality, and taste.
(And, of course, the ever-elusive ability to actually execute that I also don't think LLMs will help with.)
You need money to explore ideas.
Or an absurd amount of skill.
Say Amy is a great NodeJS and front end programer.
She can work on her own projects , but her kids can't eat hope and dreams.
Or she can get a job at SoulCrusher Solutions trying to maximum advertising revenue.
This is a compelling assertion. But who among us has truly original thoughts? How much new stuff can there be? If all the same-y stuff is losing value (but most stuff has value because it's FOUND not because it's unique) then isn't net value decreasing?
>The value of unoriginal thinking has gone down. Thinking which is quotidian and pedestrian has become even more worthless than it already was.
Imagine American manufacturing industry workers saying the same thing of the (at the time) soon-to-be import only products. Original thoughts are valued more than non-original ones but maybe, the market doesn't require that many original thoughts to extract max profit...
liked this one, but even this one turns to AI-sh style in the end... no escape from it any more.
Saturation of channels due to slop content - doesn't mean that created stuff is not scarce
there's more problems then ever before that need empathetic humans to solve - are you up for the challenge - or you're doing a quick cash grab
due to people using machine gun approaches - spray & pray - we haven't forgotten how scalable human touch is -- yeah at first - you've to do things the manual way - reach out have a conversation - but slowly word spreads around without you spending money on ads | content etc
Yeah 100% there's more legit stuff than ever, but there's also just more stuff than ever. So it becomes a discoverability problem.
To be honest this is what is discouraging me from writing more novels right now. The only reason I'm even considering it is because the love of the craft is hooked firmly in my gullet. Were it not for that I'd drop writing faster than a lava potato...
Have people naturally started sounding like an LLM when they write and talk? To me this article reads not fully human.
There's multiple things happening here. People are using LLMs to write, for sure. It's only natural to absorb what you consume, so as people read more LLM generated content they can unintentionally emit it.
And then there's classic confirmation bias; a lot of people wrote in dry academic prose.
I think this was already true but it is becoming a lot more obvious now and the effects of this problem are going to affect a lot more people as we see mass white collar worker unemployment. Some people think with lower costs to make software more software is possible. But you can try to get started building a product only to see an improved AI agent build out your idea cheaply, or a bigger company use capital to copy you and enter the same market, and so on. Also, individuals who are richer can take more risks - the loss of money is not existential for them.
The only real fix is incredibly heavy taxation of income and wealth for the ultra rich individuals and ultra large companies. That is what will break through the money moat and create fair competition.
Is the purpose of space exploration to build rockets? Or is the purpose to deliver things and people into orbit and to other worlds?
Is the purpose of a computer program to use processing, network and memory? Or is it to handle and manipulate information to give results which are useful to people?
Now the moat of having memorized intentionally convoluted and complicated programming languages has been taken away. Exactly like the printing press removed the monopoly on information which was held exclusively by priests and monks.
When tools make a job easier, they open up new markets for people to do and sell things which were too costly to offer before. AI translation alone means that small businesses can open up several markets they didn't have any access to, broadening the number of potential customers immensely.
Everyone now have access to a tool that makes them nearly as powerful as only the most creative builders were before.
But these new builders have a tool, they don’t suddenly have a newfound creativity.
I think with time we will stop seeing what we consider AI slop, simply because we know it’s not worth sharing. Instead great creative people will share very impressive things that simply wasn’t possible to build before.
Difficulty is the only true moat. [Astronaut: always has been]
Current examples: esoteric calculations that are not public knowledge; historical data that you collected and someone else didn't; valuable proprietary data; having good taste; having insider knowledge of a niche industry; making physical things; attracting an audience.
Some things that were recently difficult are now easy, but general perception has not caught up. That means there's arbitrage—you can charge the old prices for creating a web app, but execute it in a day. But this arbitrage will not last forever; we will see downward price pressure on anything that is newly easy. So my advice is: take advantage now.
Welp guess I’m done then.
Kind of nice to know I don’t have to blame myself anymore.
People here saying creativity or having a good idea is the moat
You know that if anything you build gets traction, it'll be cloned by 100 people, right?
So where are the AI clones of MS Office, JetBrains IDEs, Whatsapp, Obsidian, Sublime Text, Claude Code, etc. so far?
It isn't the moat. But it gives you a head start to build it.
It was always about money. People never cared about new things or innovation. It was always funneled through gatekeepers. The "moat" was a talking point to VC's who are used to 90% of their ventures failing and yet was taken seriously. We seem to forget all the racist jokes about how China will just make a cheaper copy, where were talks about a "moat" then?
It still take me months to build sophisticated things with hard problems at their core.
There’s not the slightest chance an LLM or less than capable developer is whipping this stuff out in a day.
It's like.. what happens when software becomes a solved problem? Like bricks for construction. There's not 50 million typically highly paid brick design engineers in the world.
I'm no big fan of economists - but if some type of business is seen as highly desirable to be in, and has minimal barriers to entry, then the market will soon be saturated. Expecting otherwise is (at best) wishful thinking.
No, it's imagination, same as it ever was.
Software to most of this discussion is a web app with a landing page, a pricing tier, and MRR. That's it. The frame is "product," the metric is "traction," and the canvas is "things people pay $9/month for."
But software is instructions that make matter and energy do things they wouldn't otherwise do. It's the most general-purpose tool humans have ever built. So let's actually think about what's underbuilt:
The whole damn physical world is barely instrumented. Agricultural systems, water infrastructure, building envelopes, soil health, local microclimates. There are farmers making irrigation decisions on vibes. Municipal water systems with no real-time leak detection. Buildings hemorrhaging energy because nobody's modeled their thermal behavior. These aren't apps. They're control systems, and they're mostly missing.
Fabrication and manufacturing are being transformed by CNC/3D printing but the software for designing things to be manufactured is still terrible (and inaccessible!). Generative design that accounts for material properties, toolpath constraints, assembly sequences. CAM software is where word processors were in 1985.
Scientific instruments. A spectrometer is mostly software now. So is a radio telescope. So is a seismograph. Every goddamn thing can be a thermometer (accidentaly!) The gap between "data sensor exists" and "useful scientific instrument" is almost entirely software, and most of that software is written by grad students in unmaintained Python.
Preservation. Some people are doing this with datamuseum.dk. But expand it: there are entire musical traditions, oral histories, craft techniques, ecological knowledge systems that exist in living memory and nowhere else. Software for capturing, encoding, and transmitting that knowledge barely exists. Not "an app for recording grandma," but formal knowledge representation of, say, how a master boatbuilder in Kerala selects wood by sound and flex.
Prosthetics and rehabilitation. This one is big for me personally! The gap between what a modern prosthetic limb could do with good software and what it actually does is enormous. Why are my eyeballs still chunks of plastic? Same for cognitive rehabilitation tools, speech therapy systems, physical therapy feedback loops.
Governance and collective decision-making. Every organization above 20 people is making decisions with tools that are basically "email plus meetings plus a shared doc." Formal deliberation systems, preference aggregation, transparent resource allocation. These are hard computer science problems that nobody's building because they don't have obvious MRR.
Tools for thought that aren't note-taking apps with backlinks. Actual reasoning aids. Argument mapping. Assumption tracking. Decision support that makes your thinking better rather than your typing faster.
The entire domain of formal verification applied to things that matter. Bridges, medical devices, voting systems, financial settlement. We have figured out how to prove some? software correct. We almost never do it for the software where correctness actually matters.
And that's me, one person, in five minutes. Every domain expert in the world is sitting on a pile of unsolved problems that software could address, and most of them have never talked to a programmer because programmers are busy building the next task management app.
Go talk to a nurse, a farmer, a building inspector, a food bank logistics coordinator. Ask them what's broken. I promise the answer isn't "nothing" and I promise nobody on ProductHunt is solving it.
PG wrote essays about this ffs! "Make something people want." "Live in the future and build what's missing." That advice didn't stop being true because AI made the building part cheaper. If anything it's more true now, because the building is almost free, which means the noticing is almost the entire game. You are skipping the noticing and going straight to the building, then wondering why nobody cares.
The number of hard things isn't going down. This thread can't see them because it's not looking at the world. It's looking at ProductHunt.
To make things concrete, in the last week I have been working on my open source speech synthesizer, rebuilding Klatt's ideas from the 1980 paper up to modern emotion/prosody work. Did you know the whole field went nuts for neural approaches in ~2018 and there's a whole shitload of interesting papers just sitting out there that nobody has ever implemented in a real system? Did you know that a bunch of people did research into what different human emotions sound like and now I can make a depressed speech synthesizer, or, scarily, one which sounds more honest to people?
Doom and gloom nonsense.
One thing I've noticed is that there is a very specific set of technologies that attract a very specific set of people. You may have seen these people jump from Bitcoin to altcoins to Ethereum to NFTs; and now to AI. The common thread behind all of these technologies is that they are inherently dehumanizing: as in, they are a means by which ambulatory piles of money can shed the skin they are ordinarily forced to wear and just exist.
AI seems like the odd man out of the group, until you understand the utter horror that is weaponized post-scarcity economics. "The only moat left is money" is the plan. It was always the plan. The goal of AI - or at least, the goal of AI to the cult of people who mindlessly agree with it is to replace humans with pliant digital slaves.
Now you have to make something actually new.
Capitalism working as intended, shifting more and more resources to the already-rich.
A handful of people doesn't own most of the country by accident.
Wealth is not a zero sum game. There’s more wealth in the world today than 50 years ago, 500 years ago, 5000 years ago.
Power is a zero-sum game. We all live in the same space, exist in the same media ecosystem, and have the same number of hours in the day. The media consumed by the public, the structure of our cities, the investments we make in science, the sources we use for energy, the laws we pass and the way they are interpreted, stewardship of our environment, and the degree to which our tax money is used to drop bombs on civilians are ALL zero-sum.
I am so sick of the pretense that GDP growth means inequality is somehow illusory. Yes, TVs used to be expensive and now they are cheap. Big whoop.
A free market is not a zero sum game indeed. Monopolies turn it into a zero sum game. To keep a market free, you have to regulate it.
Wealth concentration is dangerous for democracy, the markets and society.
I think you have it backwards. Regulation is what enables monopolies. There is no monopoly for any of the major industries like cow herding or cellular telephone service in Somalia despite almost no effective regulation. There is not even a monopoly of pirates despite them willing to use violence to try and enforce a monopoly.
If you look at the history of the US, for instance, railroad regulation was brought forth largely by the railroads because they found it impossible to form a cartel to keep up prices (due to "secret" discounting) so instead they created regulation that outlawed the kind of discounting that breaks cartels apart. A similar thing happened in banking where the banks asked for a central bank to cartelize the interest ranks to stabilize their oligopoly. And the same in pharma industry -- big pharma loves high regulatory barriers because it keeps competitors out.
A large portion of the regulation in the US was brought about as regulatory capture by corporations to increase the monopolizing effects and destroy the free market.
> Regulation is what enables monopolies
That's patently false. AT&T was a monopoly and they were broken up by antitrust regulation. The absolute most you can say is that some regulations enable monopoly. I contend that we simply should pass the good kind of regulations instead.
Monopoly is enabled by market forces such as economies of scale. Monopolization is a natural market process which happens on its own unless it is actively prevented.
> big pharma loves high regulatory barriers because it keeps competitors out
The FDA, for all the flaws of its current incarnation, is the archetype of necessary regulation. Pre-FDA, the free market did nothing whatsoever to prevent nauseating practices like the adulteration of milk with powdered plaster, lead, and cow brains. The history there is fun but quite gross.
> Somalia
What is notable about Somalia is not its lack of regulation, but the fact that it is perhaps THE least stable country on the planet. It is not the basis for any useful comparison here.
>> Regulation is what enables monopolies
>That's patently false. AT&T was a monopoly and they were broken up by antitrust regulation.
This is patently false in the context of the reply you have made -- after the invention of the telephone more and more and eventually hundreds of telephone services popped up. Then in 1918 (circa WWI), the government effectively quasi-nationalized AT&T by controlling it via a commission and the postmaster general and then AT&T leveraged politicians to create "universal telephone service" provided by AT&T and regulate competitors out of the market while using regulatory capture to use commissions to regulate rates, effectively creating a cartel that drove competitors out of business via regulation.
the whole idea of a "natural market process" here is absolute and utter hogwash. The majority of the market was AT&T competitor up until the regulators stepped in and turned it into an unnatural monopoly enforced by regulatory capture.
>The FDA, for all the flaws of its current incarnation, is the archetype of necessary regulation. Pre-FDA, the free market did nothing whatsoever to prevent nauseating practices like the adulteration of milk with powdered plaster, lead, and cow brains. The history there is fun but quite gross.
You're now arguing why we need regulation rather than whether they create monopolies or not. I see this as a complete red herring, although an interesting topic, that there are some counterpoints to.
> What is notable about Somalia is not its lack of regulation, but the fact that it is perhaps THE least stable country on the planet. It is not the basis for any useful comparison here.
What is notable is that the whole thesis is without regulation it turns into this monopolized hellscape and every inspection of that theory turns out to be false, and sometimes even the opposite.
> the whole idea of a "natural market process" here is absolute and utter hogwash
Might I introduce you to the concept of a "natural monopoly"? https://en.wikipedia.org/wiki/Natural_monopoly
Anyway, what regulation is responsible for Walmart and Amazon putting local retailers out of business?
> the government effectively quasi-nationalized AT&T
After a big merger put AT&T in charge of the majority of telephone lines in the US, the company used its control over infrastructure to drive its competitors out of business and increase its portfolio. The Justice Department tried to break up AT&T but failed; it was in the settlement of this case that AT&T was first federally regulated in 1913. Yes, AT&T's monopoly grew between 1913 and 1982, but your causality is backwards. They regulated it because it was already a monopoly.
>Anyway, what regulation is responsible for Walmart and Amazon putting local retailers out of business?
Walmart + Amazon combined are only ~16% of the retail business. They're not monopolies. The fact they put a small minority of businesses out of business does not mean they're a monopoly. This is likely part efficiencies and also part regulatory capture via the insane zoning/building regulations in this country and tax breaks that can favor large corporations.
> After a big merger put AT&T in charge of the majority of telephone lines in the US, the company used its control over infrastructure to drive its competitors out of business and increase its portfolio. The Justice Department tried to break up AT&T but failed; it was in the settlement of this case that AT&T was first federally regulated in 1913. Yes, AT&T's monopoly grew between 1913 and 1982, but your causality is backwards. They regulated it because it was already a monopoly.
... It was not already a monopoly. Hundreds of phone companies emerged and by shortly before your noted date of "regulation" those competitors held the majority market share. It became a "monopoly" after the government literally quasi-nationalized them (AT&T) to the point the fucking Postmaster General was basically in charge of it, they became intertwined with regulators, and then the drive for "universal telephone service" and regulatory commissions ensured the regulatory compliance pushed exactly into AT&Ts business model. AT&T intertwined lawmakers even brought in economics quacks to talk up natural monopolies to argue for policies to create the regulations that made AT&T a monopoly. So you have it backwards -- the regulated it from a minority market holder to an unnatural monopoly and lawmakers created this monopoly under the auspices they essentially needed to legislate a "natural" (misnomer) monopoly into existence.
> Walmart + Amazon combined are only ~16% of the retail business.
of ALL RETAIL? That includes groceries! That's huge! Anyway, Amazon is >40% of e-commerce & Walmart is >10%. Together they control more than half of all online commerce. That's definitely monopolistic.
> It was not already a monopoly. […] shortly before your noted date of "regulation" those competitors held the majority market share
From Wikipedia: "AT&T controlled over 80% of the U.S. telephone system market by 1907 and Theodore Newton Vail rejoined the company as its President. Vail negotiated with competitors, charging them fees for connecting to AT&T's long-distance network. These practices led the Justice Department to attempt to breakup AT&T, but a settlement was reached through the Kingsbury Commitment on December 13, 1913. It brought federal oversight into AT&T and led its Bell System monopoly to become federally regulated."
They had an 80% market share pre-regulation. Yes, it was already a monopoly.
This is not a personal opinion of mine, it is pretty much established science. I think only think-tank backed sources would claim the opposite.
One should understand the phenomenon as a common pattern of dynamics in unregulated markets. Not every snapshot will showcase an end state of monopolist dominated markets.
You bring up a valid point though. Regulatory capture is a indeed a weapon in the hands of anti-competitive players to prevent incumbents. Good policy usually applies differently to different strata: the small players are exempt from certain rules, or have to deal with less stringent ones than big players do, to prevent killing the market. At the risk of sounding like an llm: it is not just about policy, it is about good policy.
> major industries like cow herding
The food industry is filled with regional monopolies.
> cellular telephone service in Somalia
Ah yes, excellent example, all you have to do is completely destabilize your nation and you too can have free market capitalism.
Investors love monopolies, they fix prices and profits so their investments are not at risk. Investors hate too much competition, it lowers profits and puts their investments at risk.
Free markets need investors. Investors hate free markets. I hope you see the problem here.
> There is no monopoly for any of the major industries like cow herding or cellular telephone service in Somalia
Kinda a terrible example, as cellular telephones are highly regulated in Somalia: https://en.wikipedia.org/wiki/National_Communication_Authori...
This is a crude but misguided attempt to bypass what I say which was "effective regulation." FGS only controls a minority of Somalia -- most of it is controlled by Somaliland, Puntland, Al-Shabaab, directly by xeer (customary) law, or only FGS hanging on by a thread.
Even in cases where FGS is in control -- even then xeer law on property rights and law often supersede written law of FGS. For instance, on occasion Somali population has straight up killed FGS soldiers/police and Somalian government has deferred to xeer courts and said "welp that is fine." Xeer law in particular has a very liberal free-market view on inter-familial entrepreneurism (although with a lot of intrafamilial and tribal dues which hinder it in practice) which is at odds with what outside law has tried to impose upon it.
NCA has rather limited influence in somalia, and definitely not of the sort that could break up a monopoly if one existed.
Wealth is intently relative. Absolute growth of wealth is just inflation.
There are a lot of zero-sum things that are being chased by money, and maybe even the most sought after ones. Like political power, sexual partners, and chasing the top % of credentials for your offspring to position them with better access to zero sum things.
That's why we cured hunger in 2012, cured poverty in 2017, fixed healthcare in 2020, working as intended indeed
You might not be living in the same reality as me if you believe any of those claims.
That's called sarcasm
Had us in the first half. And the second half too.
So this is what the VCs were screaming about this bullshit about "abundance".
Abundance of copy cats that cannot make any money as prices are raced to zero.