Wow, this looks pretty cool! It seems you have quite a few features going... The main reason I haven't gotten a smartwatch yet is because of privacy concerns. But now seems like a good time to experiment with one of the watches you support.
I also still dream of one day daily driving a Linux smartphone, but that feels a bit more unrealistic to me, as I have more expectations from a phone, like being able to use bank apps and having a good battery life. But for a smartwatch, which I only expect to show me some biometrics and pass notifications from my phone, this seems perfect.
On this note: aren't JavaScript and QML/Qt too heavy/bloated for a device so small? I expect them to constrain performance and battery life quite a bit, but I admit I don't have a clue and would love to be proven wrong...
QML runs interpreted at runtime and Qt is very close to native C++, so both are actually quite lean and fast on these devices — the overhead is surprisingly small compared to full Android Wear, and we've tuned it heavily over the years to keep battery and UI responsiveness solid.
If you see devices lagging, like the sawfish (Huawei Watch 2), its usually due to updating the android base to a newer one as the manufacturer planned for and losing some of those optimizations in the process.
Wild to see such fragmentation in such a niche space. It's an aftermarket Linux flash for smartwatches, and there are companion apps for SailfishOS and Ubuntu Touch, which are extremely niche flavors of the already very niche mobile Linux.
Not being much of a watch person let alone a smartwatch aficionado, I had no idea there were even that many smart watches. The long list looks impressive. I wonder if there are a lot of the same guts so it's not as bad of a nightmare to maintain as it looks. Either way, the list of supported devices is impressive.
Indeed quite some watches share the same platform as can be seen in this list:
https://wiki.asteroidos.org/index.php/Technical_Details_of_A...
But the manufacturers all cook their own soups and its surprising how much adaption is still necessary per device.
Think of the space as less "I want Linux on my wrist", and more "I want a [cheap || not 1st world expensive] smartwatch as a gift."
These folks do gods work of making them supported and a real shared platform (c.f. their self-post "The only real signal we get is occasional [chat visitor] going "hey, it booted on my watch from 2014 and now it's usable again"")
I am running HarmonyOS on my watch (moved from Apple Watch to Huawei GT-6 41mm) now. Turns out it doesn't matter as third party apps on watches generally suck anyway (low functionality, mostly just companions to iOS/Android app, bad/not many options for UI).
Thanks for sharing, indeed it is good to reuse and improve hardware whenever possible rather than just going to waste. Had never heard of this project before, so it was a good find.
I see the supported watches, still have a doubt: what is scenario looking like for the cheap smarwatches from Aliexpress and similar? Any chance to also flash a new firmware or is this limited to more CPU-powerful watches?
These are all Linux kernel-based WearOS watches (not just smartbands running a barebones microcontroller), so could they be running a mainlined kernel and Linux OS such as pmOS? Of course the UI layer might be specific to the form factor, but everything else could just be standard.
In theory yes. Asteroidos has experimental support for a mainline watch.
Most vendors don't upstream their drivers and kernel mods. Also Android drivers use an abstraction layer and a different format to some extentent. So you have to reverse and write your own driver.
So they could run mainline if the vendor or a user bothers to upstream drivers and hardware quirks.
A lot of the vendors don't meet quality expectations of the kernel team and sources are usually for older kernel versions and the code would need changes or refactoring.
The mainlining work is usually done by the community, not necessarily vendors. The relevant pmOS wiki page is https://wiki.postmarketos.org/wiki/Category:Watch but it may be less comprehensive or up-to-date than the AsteroidOS hardware support list. Everyone can help by adding technical info about hardware to the wiki!
This is great! I'm going to install it immediately. I have a Fossil 6. It's in my opinion the best looking smartwatch yet released however the WatchOS performance and functionality is absolutely dreadful. I'm excited to bring this watch back from the dead!
Hey, thanks for the new release. I should definitely fix my wristband and start wearing my AsteroisOS watch again (LG Lenok).
You have probably addressed that somewhere, but would it be possible to run your UI stack somewhere else? (PostmarketOS).
My other wish for AsteroidOS would be for it to leverage Wi-Fi better. Not sure how much more energy it would use, but having a longer range for my notifications would be nice (at least on LAN). Being able to perform a few other actions independently of my phone would be great: weather % time updates, e-mail notifications, home assistant control, etc. I get that it may affect battery life as well.
While I'm at it: tiny bug report, but I adjusted the time while the stopwatch was running, and this affected the stopwatch result.
We have implemented a wifi toggle in the quickpanel with 2.0. But the wifi credentials still need to be entered into connmanctl on the cli. As soon as you got wifi set up and connected, you can already now sync weather data usin asteroid-weatherfetch. But right, wifi usually uses up to 30% more power and should be enabled selectively.
For the postmarket question, yes, it is our longterm goal to mainline watches, which we are sort of doing in coorperation with the postmarket guys. But thats a humongous task and part of the idea of this 2.0 release is to interest capable contributors to push things further ;)
This looks cool!! I have an Apple Watch that I’ve been perfectly satisfied with, but it strikes me that AsteroidOS might allow me to explore other unique smart devices relatively cheaply, by buying used/non-supported stuff and trying it out. Are there any in particular that you’d recommend to a newbie? Hopefully something that is both eBay-able[1] and fun?
I've always wanted a watch whose face colors can be changed to fit my outfit. It seems this is doable albeit at the cost of battery but I havent seen any software that allows it
This is totally doable! The Always on Display screen can have colors and those do not impact the battery life much. Nothing hinders you to show a full color as AoD. There are some watches however that only support 4 colors in AoD mode. Namely the ASUS ones. If you are serious about developing such a watchface, come to our matrix chat for more tips where to start.
This amazing! I am definitely a target audience for this (I don't run any phone without at least unlocked bootloader, root and lsposed). So happy to see ROMs are coming to Android watches!
This is an awesome project. Props to y'all for just making something you want to exist!
I have a Tizen-based Samsung watch (Gear Sport, 2017). It's served me faithfully but I'm starting to notice the battery degrading. I'd be interested in trying AsteroidOS with it, if Tizen support ever lands.
Thats awesome! Recently I was looking into making apps for my smartwatch that don't exist (like watch display with multiple timezones), and infrastructure to make your own apps is very poor.
One thing I wish for is Rust support, since its running Linux it should be possible, isn't it?
It would be possible to use Rust. Nobody got around working on it tbh. But simple things like your mentioned watchface idea are really quick to do in QML.
Can anyone suggest where to find a watch that is supported if you live in the US? I've been scanning eBay but it feels difficult to get ahold of a supported device. Are there sites that ship to the US where a new or used device can be found?
Usually the Ticwatch Pro 2018/2020 (catfish) is widely available since it was a popular model. The more recent version Ticwatch Pro 3 (rubyfish/rover) is freshly ported and not as well supported as the first Ticwatch Pro yet. I bought one new in box just this week from german ebay for 70€.
We got a Team member in the US/ML who is hoarding watches and seems to have no problem acquiring them :D I wish you luck.
While the PineTime/Infinitime is an awesome project everyone in our community roots for, the hardware platforms are totally different. The PineTime is microcontroller based and works on just 64KB memory and 512KB flash. AsteroidOS is a linux distro tailored for devices with SoC, half a gig of memory and >2GB storage.
Arguably its much smarter on many levels to use the microcontroller approach for the rather simple smartwatch task. But somehow these full SoC general purpose computing smartwatches have been produced and we as linux people slipped into the niche of seizing those.
I would love if it would support some of the no-brand Chinese watches you get usually for cheap, the hardware is great but the software usually is bad or outdated. I use one now, I don’t even know anything about it other than the Bluetooth name and app name, but it’s good in measuring distance, blood pressure, heart rate, sleep, among others that’s surprisingly it’s very accurate, it also has a builtin strong flashlight, I like it but with a fully fledged linux would definitely be better.
I have a galaxy watch 4 which I'd hoped was old-enough to be supported but I can see that it is not. I get it, hardware is hard.
I'm curious, is the challenge with newer hardware lack of chipset drivers for modern watches, or is there a fundamental difference between the new devices and the old ones that make them completely incompatible with asteroidOS?
With the latest devices its usually a problem that manufacturers choose to omit the usb pin outs in favor of water resistance and wireless charging. Making them a challenge to flash and still wear afterwards.
Another issue is that we currently need to rely on libhybris for quick porting process that employs the android drivers. And the new devices run android versions that libhybris can not handle yet. Its just a slow process on all fronts but we are actually releasing this 2.0 publicly to possibly interest more developers.
The main UI and GUI components are Qt. So you could use Qt bindings to build something with Rust.
If you don't want the same look and feel, it's just a normal linux with wayland and systemd. Cross compile to the architecture and adapt the UI to the small displays and you should be fine.
Perfect! Our Ring is libhybris, Sauron is the binary blob drivers the manufacturers cursed us with, and we're still hobbling toward mainline Mordor one old kernel backport at a time. One does not simply upstream a smartwatch driver.
Fair suspicion in 2026 — but we actually started drafting that announcement in early 2023 before first committing it here: https://github.com/jrtberlin/aos2.0-post. As a non native speaker i sure ran my commits through grammar polishing multiple times over the years. And i am genuinely curious now what to avoid to not sound like an LLM if you could dig out one or two examples.
I wonder if they meant this HN post (the OP), rather than https://asteroidos.org/news/2-0-release/? The latter doesn't seem LLM-written to me, but the HN post does.
Full disclosure, i spent ~3 hours crafting this post to hit the tone i wanted to convey and am kind of proud of the wrist-size linux banger i came up with. I am usually not good with writing since it takes me ages and i might be a bit over sensitive right now. All i wanted was to spare you all the rocky grammar as a cherry on top. To now find that the polished version triggers your "its (completely?) Ai written" sensor. Lesson learned i guess.
FWIW as someone who sadly has about 10% of his brain constantly engaged in AI detection when I read HN, absolutely nothing about the announcement struck me as feeling AI generated.
(I feel guilty for pointing it out but I guess it's actually a compliment in this context: I even saw a beautifully reassuring mis-conjugated verb elsewhere on the website. I wonder if LLMs will have to start injecting these errors to give us an authentic feel. Maybe they already have).
I think we are at a point now where just coz AI is so prevalent, every post on any programming forum will have at least one comment saying "AI slop".
I hadn't opened the article yet and was just browsing comments over my cereal when I saw this and thought "ugh, amother one?" and went to check for myself
I didn't get an LLM vibe at all. Looking for it specifically, the bullet point about UI improvements is a candidate; the sentence following "mediawiki" could be an autocompletion; maybe the first sentence of the download section... but they're also all plausibly just a bit 'marketing team worded', so not necessarily LLM-sounding. And even if an LLM made suggestions to some small parts like these, who cares? There aren't any slop sections that waste your time, this is just like using a thesaurus — if these parts were LLM suggestions in the first place, which I don't actually expect because then there should be more of it
This type of post very poorly lends itself for auto-writing anyway. It'll put emphasis on the wrong aspects and not come out as intended, at least in my experience it's more work coaxing it to good results. It can be helpful not to start from a blank page but that's about it, I rarely find a sentence among the output that's fully usable as-is
I was referring to the text of the HN post. The second and third paragraphs in particular, but most of it, really. If you didn’t use an LLM for that then congratulations, you’ve successfully nailed the Silicon Valley LinkedIn marketing guru zeitgeist which LLMs seem to gravitate towards.
Not on this article, but everywhere nowadays are hints of AI. For example using emoticons on the beginning of each sentence or never really going deep into the topic. The complete lack of grammatical errors is another tip, unless the prompter asks to write minor errors and pretend to be a human.
Wouldn't be suprised that before the end of this year we see more website declaring themselves to be "human-only zones", considering we won't be able to match the speed and quantity and replies from bots. Making it difficult to hold proper conversations.
I don't care if the text is ai-written as long as it delivers the message that it is supposed to deliver correctly. If none of the engineers on the team feels like writing long announcement posts - why couldn't they use ai, check/correct its output and go back enjoy hacking smartwatches?
I think people are getting too fixated on how exactly each symbol was created instead of what the message is.
There is misunderstanding. There is nothing against using it for those cases, the problem is when it gets used for producing long and unverified slop or debatting against humans.
This is one theory about the controversial age identification currently being implemented happily by social media companies: it's actually about human identification. That's the new captcha.
Wow, this looks pretty cool! It seems you have quite a few features going... The main reason I haven't gotten a smartwatch yet is because of privacy concerns. But now seems like a good time to experiment with one of the watches you support.
I also still dream of one day daily driving a Linux smartphone, but that feels a bit more unrealistic to me, as I have more expectations from a phone, like being able to use bank apps and having a good battery life. But for a smartwatch, which I only expect to show me some biometrics and pass notifications from my phone, this seems perfect.
On this note: aren't JavaScript and QML/Qt too heavy/bloated for a device so small? I expect them to constrain performance and battery life quite a bit, but I admit I don't have a clue and would love to be proven wrong...
> The main reason I haven't gotten a smartwatch yet is because of privacy concerns
Have you looked into Gadgetbridge? https://gadgetbridge.org/gadgets/
QML runs interpreted at runtime and Qt is very close to native C++, so both are actually quite lean and fast on these devices — the overhead is surprisingly small compared to full Android Wear, and we've tuned it heavily over the years to keep battery and UI responsiveness solid. If you see devices lagging, like the sawfish (Huawei Watch 2), its usually due to updating the android base to a newer one as the manufacturer planned for and losing some of those optimizations in the process.
Thanks for clarifying and wish you all the best with the project! I'll see what watch I can get my hands on.
BangleJS has an embedded JS interpreter and runs just fine for days without a battery recharge.
Wild to see such fragmentation in such a niche space. It's an aftermarket Linux flash for smartwatches, and there are companion apps for SailfishOS and Ubuntu Touch, which are extremely niche flavors of the already very niche mobile Linux.
Not being much of a watch person let alone a smartwatch aficionado, I had no idea there were even that many smart watches. The long list looks impressive. I wonder if there are a lot of the same guts so it's not as bad of a nightmare to maintain as it looks. Either way, the list of supported devices is impressive.
Indeed quite some watches share the same platform as can be seen in this list: https://wiki.asteroidos.org/index.php/Technical_Details_of_A... But the manufacturers all cook their own soups and its surprising how much adaption is still necessary per device.
This is a feature, not a bug. Like the many Linux distros we have or the many UI/WM you can use on desktop Linux.
> niche space
Think of the space as less "I want Linux on my wrist", and more "I want a [cheap || not 1st world expensive] smartwatch as a gift."
These folks do gods work of making them supported and a real shared platform (c.f. their self-post "The only real signal we get is occasional [chat visitor] going "hey, it booted on my watch from 2014 and now it's usable again"")
I am running HarmonyOS on my watch (moved from Apple Watch to Huawei GT-6 41mm) now. Turns out it doesn't matter as third party apps on watches generally suck anyway (low functionality, mostly just companions to iOS/Android app, bad/not many options for UI).
Thanks for sharing, indeed it is good to reuse and improve hardware whenever possible rather than just going to waste. Had never heard of this project before, so it was a good find.
I see the supported watches, still have a doubt: what is scenario looking like for the cheap smarwatches from Aliexpress and similar? Any chance to also flash a new firmware or is this limited to more CPU-powerful watches?
Most cheap watches are microcontroller-only. AsteroidOS is a Linux distro and therefore needs a chip capable of running some Linux variant.
Initially, I read Asteroids 2 :P If someone clicked like me thinking of the game, I recommend playing this one :D https://coderai.itch.io/clashroids
Turns out you can play Asteroids on AsteroidOS https://github.com/moWerk/asteroid-blaster
These are all Linux kernel-based WearOS watches (not just smartbands running a barebones microcontroller), so could they be running a mainlined kernel and Linux OS such as pmOS? Of course the UI layer might be specific to the form factor, but everything else could just be standard.
In theory yes. Asteroidos has experimental support for a mainline watch. Most vendors don't upstream their drivers and kernel mods. Also Android drivers use an abstraction layer and a different format to some extentent. So you have to reverse and write your own driver.
So they could run mainline if the vendor or a user bothers to upstream drivers and hardware quirks.
A lot of the vendors don't meet quality expectations of the kernel team and sources are usually for older kernel versions and the code would need changes or refactoring.
The mainlining work is usually done by the community, not necessarily vendors. The relevant pmOS wiki page is https://wiki.postmarketos.org/wiki/Category:Watch but it may be less comprehensive or up-to-date than the AsteroidOS hardware support list. Everyone can help by adding technical info about hardware to the wiki!
This is seriously impressive! Never knew after market os's were even a thing for watches with their proprietary drivers.
I like that peeking watch face switcher, companies like samsung even after all these years still takes way too long to apply a watch face.
Thanks so much! The peek gesture is inherited from lipstick and we kind of built our UI around those possibilities.
This is great! I'm going to install it immediately. I have a Fossil 6. It's in my opinion the best looking smartwatch yet released however the WatchOS performance and functionality is absolutely dreadful. I'm excited to bring this watch back from the dead!
Hey, thanks for the new release. I should definitely fix my wristband and start wearing my AsteroisOS watch again (LG Lenok).
You have probably addressed that somewhere, but would it be possible to run your UI stack somewhere else? (PostmarketOS).
My other wish for AsteroidOS would be for it to leverage Wi-Fi better. Not sure how much more energy it would use, but having a longer range for my notifications would be nice (at least on LAN). Being able to perform a few other actions independently of my phone would be great: weather % time updates, e-mail notifications, home assistant control, etc. I get that it may affect battery life as well.
While I'm at it: tiny bug report, but I adjusted the time while the stopwatch was running, and this affected the stopwatch result.
Nice, thanks for the bug report! I have made in issue in the stopwatch repo: https://github.com/AsteroidOS/asteroid-stopwatch/issues/13
We have implemented a wifi toggle in the quickpanel with 2.0. But the wifi credentials still need to be entered into connmanctl on the cli. As soon as you got wifi set up and connected, you can already now sync weather data usin asteroid-weatherfetch. But right, wifi usually uses up to 30% more power and should be enabled selectively.
For the postmarket question, yes, it is our longterm goal to mainline watches, which we are sort of doing in coorperation with the postmarket guys. But thats a humongous task and part of the idea of this 2.0 release is to interest capable contributors to push things further ;)
This looks cool!! I have an Apple Watch that I’ve been perfectly satisfied with, but it strikes me that AsteroidOS might allow me to explore other unique smart devices relatively cheaply, by buying used/non-supported stuff and trying it out. Are there any in particular that you’d recommend to a newbie? Hopefully something that is both eBay-able[1] and fun?
[1] of course open to other sources as well
I've always wanted a watch whose face colors can be changed to fit my outfit. It seems this is doable albeit at the cost of battery but I havent seen any software that allows it
This is totally doable! The Always on Display screen can have colors and those do not impact the battery life much. Nothing hinders you to show a full color as AoD. There are some watches however that only support 4 colors in AoD mode. Namely the ASUS ones. If you are serious about developing such a watchface, come to our matrix chat for more tips where to start.
This amazing! I am definitely a target audience for this (I don't run any phone without at least unlocked bootloader, root and lsposed). So happy to see ROMs are coming to Android watches!
This is an awesome project. Props to y'all for just making something you want to exist!
I have a Tizen-based Samsung watch (Gear Sport, 2017). It's served me faithfully but I'm starting to notice the battery degrading. I'd be interested in trying AsteroidOS with it, if Tizen support ever lands.
Thats awesome! Recently I was looking into making apps for my smartwatch that don't exist (like watch display with multiple timezones), and infrastructure to make your own apps is very poor.
One thing I wish for is Rust support, since its running Linux it should be possible, isn't it?
It would be possible to use Rust. Nobody got around working on it tbh. But simple things like your mentioned watchface idea are really quick to do in QML.
> wrist-sized Linux
What a charming turn of phrase!
we're finally upon the year of the linux wristwatch
Can anyone suggest where to find a watch that is supported if you live in the US? I've been scanning eBay but it feels difficult to get ahold of a supported device. Are there sites that ship to the US where a new or used device can be found?
Usually the Ticwatch Pro 2018/2020 (catfish) is widely available since it was a popular model. The more recent version Ticwatch Pro 3 (rubyfish/rover) is freshly ported and not as well supported as the first Ticwatch Pro yet. I bought one new in box just this week from german ebay for 70€. We got a Team member in the US/ML who is hoarding watches and seems to have no problem acquiring them :D I wish you luck.
This sounds like an interesting investigatory problem you've put in front of me...
Thanks so much!
I can see some Fossil Gen 6's on eBay in the US. (I only learned about AsteroidOS today and I'm not in the US)
Did you consider PineTime?
While the PineTime/Infinitime is an awesome project everyone in our community roots for, the hardware platforms are totally different. The PineTime is microcontroller based and works on just 64KB memory and 512KB flash. AsteroidOS is a linux distro tailored for devices with SoC, half a gig of memory and >2GB storage. Arguably its much smarter on many levels to use the microcontroller approach for the rather simple smartwatch task. But somehow these full SoC general purpose computing smartwatches have been produced and we as linux people slipped into the niche of seizing those.
This makes me want to go find some old watches! Love to commitment to privacy and longevity.
Great work to everyone involved with the project!
Thanks for sharing this, it looks promising!
I would love if it would support some of the no-brand Chinese watches you get usually for cheap, the hardware is great but the software usually is bad or outdated. I use one now, I don’t even know anything about it other than the Bluetooth name and app name, but it’s good in measuring distance, blood pressure, heart rate, sleep, among others that’s surprisingly it’s very accurate, it also has a builtin strong flashlight, I like it but with a fully fledged linux would definitely be better.
Heh, I made myself laugh.
Literally clicked on that link thinking I had read _asteroids_, as in a remake of the 1979 game.
Total disappointment ensued.
I have a galaxy watch 4 which I'd hoped was old-enough to be supported but I can see that it is not. I get it, hardware is hard.
I'm curious, is the challenge with newer hardware lack of chipset drivers for modern watches, or is there a fundamental difference between the new devices and the old ones that make them completely incompatible with asteroidOS?
With the latest devices its usually a problem that manufacturers choose to omit the usb pin outs in favor of water resistance and wireless charging. Making them a challenge to flash and still wear afterwards. Another issue is that we currently need to rely on libhybris for quick porting process that employs the android drivers. And the new devices run android versions that libhybris can not handle yet. Its just a slow process on all fronts but we are actually releasing this 2.0 publicly to possibly interest more developers.
well done!
Rust support?
The main UI and GUI components are Qt. So you could use Qt bindings to build something with Rust. If you don't want the same look and feel, it's just a normal linux with wayland and systemd. Cross compile to the architecture and adapt the UI to the small displays and you should be fine.
Damn, not even LLM copy. Frodo lives.
Perfect! Our Ring is libhybris, Sauron is the binary blob drivers the manufacturers cursed us with, and we're still hobbling toward mainline Mordor one old kernel backport at a time. One does not simply upstream a smartwatch driver.
fyi the announcement is very obviously ai-written. It’s off-putting.
Fair suspicion in 2026 — but we actually started drafting that announcement in early 2023 before first committing it here: https://github.com/jrtberlin/aos2.0-post. As a non native speaker i sure ran my commits through grammar polishing multiple times over the years. And i am genuinely curious now what to avoid to not sound like an LLM if you could dig out one or two examples.
I wonder if they meant this HN post (the OP), rather than https://asteroidos.org/news/2-0-release/? The latter doesn't seem LLM-written to me, but the HN post does.
Correct.
Full disclosure, i spent ~3 hours crafting this post to hit the tone i wanted to convey and am kind of proud of the wrist-size linux banger i came up with. I am usually not good with writing since it takes me ages and i might be a bit over sensitive right now. All i wanted was to spare you all the rocky grammar as a cherry on top. To now find that the polished version triggers your "its (completely?) Ai written" sensor. Lesson learned i guess.
FWIW as someone who sadly has about 10% of his brain constantly engaged in AI detection when I read HN, absolutely nothing about the announcement struck me as feeling AI generated.
(I feel guilty for pointing it out but I guess it's actually a compliment in this context: I even saw a beautifully reassuring mis-conjugated verb elsewhere on the website. I wonder if LLMs will have to start injecting these errors to give us an authentic feel. Maybe they already have).
I think we are at a point now where just coz AI is so prevalent, every post on any programming forum will have at least one comment saying "AI slop".
Read the text of the HN post. If it’s not AI slop, then it’s a canonical example of the training data that produces the slop.
I hadn't opened the article yet and was just browsing comments over my cereal when I saw this and thought "ugh, amother one?" and went to check for myself
I didn't get an LLM vibe at all. Looking for it specifically, the bullet point about UI improvements is a candidate; the sentence following "mediawiki" could be an autocompletion; maybe the first sentence of the download section... but they're also all plausibly just a bit 'marketing team worded', so not necessarily LLM-sounding. And even if an LLM made suggestions to some small parts like these, who cares? There aren't any slop sections that waste your time, this is just like using a thesaurus — if these parts were LLM suggestions in the first place, which I don't actually expect because then there should be more of it
This type of post very poorly lends itself for auto-writing anyway. It'll put emphasis on the wrong aspects and not come out as intended, at least in my experience it's more work coaxing it to good results. It can be helpful not to start from a blank page but that's about it, I rarely find a sentence among the output that's fully usable as-is
I was referring to the text of the HN post. Go read it.
Aye-aye o7 ...
I saw that one already and didn't suspect slop there either
I was referring to the text of the HN post. The second and third paragraphs in particular, but most of it, really. If you didn’t use an LLM for that then congratulations, you’ve successfully nailed the Silicon Valley LinkedIn marketing guru zeitgeist which LLMs seem to gravitate towards.
What is off-putting ia this kind of comments everywhere you go.
This is the new "this has clearly been photoshoped" meme we used to see on every forum thread 2 decades ago and it is annoying as hell.
Not on this article, but everywhere nowadays are hints of AI. For example using emoticons on the beginning of each sentence or never really going deep into the topic. The complete lack of grammatical errors is another tip, unless the prompter asks to write minor errors and pretend to be a human.
Wouldn't be suprised that before the end of this year we see more website declaring themselves to be "human-only zones", considering we won't be able to match the speed and quantity and replies from bots. Making it difficult to hold proper conversations.
I don't care if the text is ai-written as long as it delivers the message that it is supposed to deliver correctly. If none of the engineers on the team feels like writing long announcement posts - why couldn't they use ai, check/correct its output and go back enjoy hacking smartwatches?
I think people are getting too fixated on how exactly each symbol was created instead of what the message is.
There is misunderstanding. There is nothing against using it for those cases, the problem is when it gets used for producing long and unverified slop or debatting against humans.
This is one theory about the controversial age identification currently being implemented happily by social media companies: it's actually about human identification. That's the new captcha.
Tough for you, because anyone with any kind of standards is likely to continue objecting to lazy, mediocre, automated posts whenever they happen.