News for hackers. Stuff that matters.
Re:AI in Linux by throwuxiytayq on Wednesday July 29, 2026 @12:06PM (#49099314)
tl;dr Kernel devs should prompt less because it uses up water or something.

Repeat after me: you can't put the cat back in the bag. Pandora's box is open. Sucks, but you have to learn to live with the reality that most developers will use AI to write code from now on.

Re:AI in Linux by NewJazz on Wednesday July 29, 2026 @12:15PM (#49099432)
[delayed]
Re:AI in Linux by berkes on Wednesday July 29, 2026 @12:30PM (#49099616)
> Pandora's box is open.

And it's not Linux' task to close that, even if it could, which it cannot. How is that disingenuous?

I agree that we should keep discussing the merits and dangers of AI-aided software development. Broad, so indeed also CO2 etc. But that discussion is not something that Linux should push forward. It's a discussion that should take place outside the context of specific projects and in the context of the AI tools and providers, IMO.

Re:AI in Linux by ch71r22 on Wednesday July 29, 2026 @12:44PM (#49099834)
It's disingenuous because it's attacking something that the article didn't say. It didn't say "kernel devs should prompt less because it uses up water."

The real argument was that AI has many externalities that should be considered (climate change, water use, energy prices, putting people out of jobs) and that Linux, as a relevant and influential political actor, should engage with these topics. It's not even giving a final determination about what Linux should do, it's just disagreeing with Torvalds' total rejection of ethical arguments about AI.

Even if you believe that the technical merits outweigh the other costs, it seems at least reasonable to engage on those grounds. But Torvalds isn't doing that.

Also, simply stating that the "cat is out of the bag" and "Pandora's box is open" isn't really an argument. These are thought-terminating cliches; not logical arguments

Re:AI in Linux by throwuxiytayq on Wednesday July 29, 2026 @01:58PM (#49100813)
> These are thought-terminating cliches; not logical arguments

Consider sometimes terminating some thoughts and replacing them with actually useful ones. It’s good for you and your goals regardless of what they are.

People burning thought-hours on pretending that we can make AI go away are basically doing nothing except emitting CO2 and consuming water. If you want change, start with yourself!

Anyway, so far zero actionable suggestions for Torvalds in the article or in this thread. Great job so far.

Re:AI in Linux by ch71r22 on Wednesday July 29, 2026 @03:53PM (#49102173)
The original discussion was based on actionable suggestions about an opt-out system for Sachiko
Re:AI in Linux by veeti on Wednesday July 29, 2026 @12:30PM (#49099630)
Contributors may object to the Hitler AI on their personal ethical grounds, but it's not going to make their buggy and vulnerable code any less so. Because our societies critically depend on projects like Linux, it follows that someone willing to pick up the pieces and do the work will then take their place.
Re:AI in Linux by kouteiheika on Wednesday July 29, 2026 @12:55PM (#49100005)
> This is a really disingenuous way to characterize the article (CO2 and global warming were much more of a focus than water, so your "or something" is doing a lot of work).

Remind me please, how much exactly do data centers contribute to greenhouse emissions? And how much something like, I don't know, meat production contribute?

If my Googling is correct, apparently a single ChatGPT query produces ~0.004 kg CO2e/kg. Beef production contributes ~60 kg CO2e/kg. So you'd need to make, err... over 15k ChatGPT queries to produce as much emissions as eating a single kilogram of beef.

I've seen anti-AI protests. Where are all the anti-meat protests? I don't mean to engage in whataboutism, I'm just saying -- if your goal is to reduce CO2 in the atmosphere then there are bigger fish to fry, instead of just virtue signaling because there's a new trendy bogeyman in town.

Re:AI in Linux by NewJazz on Wednesday July 29, 2026 @01:04PM (#49100102)
What's with this whataboutism? What if I told you I was vegan? What if I told you I only drive to work or to medical appointments? Would you pull another excuse out to justify your emissions?

I've seen anti-AI protests. Where are all the anti-meat protests?

Fucking everywhere bro.

Re:AI in Linux by aziaziazi on Wednesday July 29, 2026 @02:28PM (#49101192)
Perfect time to plug some coming events. Anybody welcome to try some free Dahl, Tofu, Tempeh and others delicacies.

  29.08.2026 end-of-speciesism.org/en/
  1.10.2026 wikipedia.org/wiki/World_Vegetarian_Day
  2.10.2026 dayforanimals.com (birthday of Gandhi)
  1.11.2026 wikipedia.org/wiki/World_Vegan_Day
Re:AI in Linux by throwuxiytayq on Wednesday July 29, 2026 @01:44PM (#49100623)
Kernel developers just aren’t emitting that much CO2 with their proompting. It’s okay that you’re pissed off, but it would be more productive if you were pissed off by real world issues.
Re:AI in Linux by TiredOfLife on Thursday July 30, 2026 @12:37AM (#49106160)
> What if I told you I was vegan?

That means you are likely responsible for more water wastage than a datacenter

Re:AI in Linux by NewJazz on Monday August 10, 2026 @01:14PM (#49246714)
How is water used to grow food "waste"?
Re:AI in Linux by YorickPeterse on Wednesday July 29, 2026 @12:23PM (#49099540)
There are many arguments one could make in favor of keeping AI around and using it, but your "argument" here is just about the lousiest and most naive argument one could possibly make, and shows a complete lack of empathy and understanding from your end.
Re:AI in Linux by throwuxiytayq on Wednesday July 29, 2026 @12:55PM (#49100007)
I'm not making an argument to keep AI around or use it, I'm just saying that it stays. Reading comprehension on the internet these days...
Re:AI in Linux by tavoe on Wednesday July 29, 2026 @12:32PM (#49099647)
How do I know if comments like this come from a real person or a bot?

Your suggestion that "anything popular cannot be morally wrong" is proposterous. It feels like you're trying to bait people into arguing with you in a way I associate with bot spam.

But on the off change you're a real person - I will politely say - just because something is useful and popular doesn't meant it doesn't have negative externalities and doesn't mean people shouldn't fight to make it better.

Improving emissions standards is important even though lots of people drive cars. The same can absolutely be true of AI.

Re:AI in Linux by keeda on Wednesday July 29, 2026 @01:27PM (#49100437)
> Your suggestion that "anything popular cannot be morally wrong" is proposterous.

It’s not about being popular, it’s about facing a new economic reality. Once machines were invented there was no going back to human labor, no matter how bad it was for the displaced workers or the environment.

Technology has always been a relentless economic force. And this is largely because despite the myriad externalities, it’s blindingly obvious that we as a species are, on average, much better off with technology than without.

Re:AI in Linux by throwuxiytayq on Wednesday July 29, 2026 @12:49PM (#49099914)
> How do I know if comments like this come from a real person or a bot?

Presumably, I'm a bot because you disagree with me.

> Your suggestion that "anything popular cannot be morally wrong" is proposterous. > just because something is useful and popular doesn't meant it doesn't have negative externalities and doesn't mean people shouldn't fight to make it better.

PREposterous. And I said nothing of the sort. It feels like you're hallucinating things nobody said in a way I strongly associate with bot spam.

> Improving emissions standards is important even though lots of people drive cars. The same can absolutely be true of AI.

The article did not at any point suggest that we need to improve and optimize AI. There is no magic wand that Linus can wave to measure how much water the incoming AI-assisted contributions are using. The guy should just go protest at a data center construction site or something; the article is pointless and completely fails to diagnose the actual problems with AI or propose viable solutions.

Re:AI in Linux by veeti on Wednesday July 29, 2026 @12:45PM (#49099849)
So, instead of getting upset at the people taking a joyride on the gas guzzling Lamborghini or generating shitty slop memes on TikTok, you're upset the ambulance drivers are racing to help the world out.
Re:AI in Linux by john_strinlai on Wednesday July 29, 2026 @12:36PM (#49099714)
if you suspect someone is a bot, email the mods.

if you're just saying it as an ad hominem, that is pretty lame, and you can make a stronger argument without it.

Re:AI in Linux by bigstrat2003 on Wednesday July 29, 2026 @12:42PM (#49099800)
> Sucks, but you have to learn to live with the reality that most developers will use AI to write code from now on.

No I don't. I can urge them to actually do their jobs and not just turn out slop, while refusing to use the slop machine myself.

Re:AI in Linux by throwuxiytayq on Wednesday July 29, 2026 @03:57PM (#49102231)
Good luck to you, but I hope you’re mentally prepared to face the future where an average software developer increases their use of AI by orders of magnitude.