this post was submitted on 29 Mar 2026
55 points (79.6% liked)

Linux

13059 readers
240 users here now

A community for everything relating to the GNU/Linux operating system (except the memes!)

Also, check out:

Original icon base courtesy of lewing@isc.tamu.edu and The GIMP

founded 2 years ago
MODERATORS
all 21 comments
sorted by: hot top controversial new old
[–] turdas@suppo.fi 42 points 2 days ago (1 children)

Getting the hottest new Linux reporting from *checks address bar* PC Gamer?

[–] Kirk@startrek.website 33 points 2 days ago

PC Gamer is just reporting on the original story from the Register, and the quote is real from the maintainer of the stable branch.

[–] artyom@piefed.social 37 points 2 days ago (3 children)

Why can't we ignore it again?

[–] Vincent@feddit.nl 8 points 2 days ago (1 children)

If these tools are indeed finding security issues, then ignoring them means someone else will find those issues - and abuse them.

[–] artyom@piefed.social -2 points 2 days ago* (last edited 2 days ago) (1 children)

Doesn't matter if they find security issues (they won't) if they're buried in a veritable haystack of false reports.

[–] Vincent@feddit.nl -1 points 2 days ago (1 children)

That's true. If they're not, though, or if they're easy to generate yourself, then you are kinda forced to pay attention though, if you care about the security of your project.

I don't have the expertise or experience to say whether that is true. But GregKH seems to think so, and other prolific projects seem to be coming to the same conclusions.

I get that it's attractive to think that AI isn't capable of it. But it's important that what you believe to be true is, and stays, based on reality rather than on what I wish is true. And it's especially important to be wary of when you really want something to be true.

[–] artyom@piefed.social 3 points 2 days ago (1 children)

I get that it's attractive to think that AI isn't capable of it

LOL you think this is just what I want to believe? Quite the opposite, I assure you.

But GregKH seems to think so, and other prolific projects seem to be coming to the same conclusions.

Lots of people are deluded, and subject to mental manipulation, unable to understand what's happening in front of them. Falling prey to powerful marketing with unlimited budgets. Ever heard of "AI psychosis"?

[–] Vincent@feddit.nl 1 points 2 days ago

Oh sorry, that might have been projecting - it's what I want to believe.

[–] lambalicious@lemmy.sdf.org 14 points 2 days ago (2 children)

Unfortunately AI has gotten ahold of several projects so it's not as easy to ignore. And with Linux itself being on the list, it seems the time comes for the community to migrate to Haiko or BSD.

[–] artyom@piefed.social 16 points 2 days ago (2 children)

Unfortunately AI has gotten ahold of several projects

Why does that matter?

[–] TacoSocks 9 points 2 days ago (1 children)
  • Legal Ramifications
  • Legal Cases And Law Problems
  • License Problems
  • Stolen Training Data
  • Environmental Impact
  • Labor
  • Poor Code Quality
  • Deskilling
  • Infosec risks
  • Healthy and Safety
  • Ties to the War Industrial Complex
  • Effects on Policing
  • Maintainer Fatigue
  • Effect on Hardware Prices

This website linked in the post you replied to lists a bunch of reasons.

[–] artyom@piefed.social 2 points 2 days ago (1 children)

Those all seem like great reasons to ignore AI...

[–] TacoSocks 2 points 2 days ago (1 children)

When you say ignore AI, do you mean stick your head in the sand and hope it goes away or actively avoid interacting with AI and AI based projects?

[–] artyom@piefed.social 3 points 2 days ago* (last edited 2 days ago)

I don't expect it to "go away". It was around long before ChatGPT and it'll be around long after it. I'm just crossing my fingers that the bubble pops sooner rather than later so we can just go back to it being just a nonsense marketing term instead of the thing people are talking about every 5 seconds, for whatever reason.

But besides, that, both. Ive been yelled at that AI is the second coming of Jesus nonstop for the last 4 years. And every time I've given it an ounce of consideration, it massively disappoints. And so now I want to stick my head in the sand until those people STFU.

I think the whole market hinges on what is essentially a "false dawn". People think "we're so close" and some innovation is right around the corner that's going to suddenly make it useful, but I don't think its coming. I think we're probably 20+ years away.

[–] lambalicious@lemmy.sdf.org 0 points 2 days ago

It reduces the foothold available for AI-free projects, in particular once "big enough" projects like Firefox or Linux get infected. Since there is significant inertia to switching to, or even developing, an alternative (a web browser might have been casual dev in 1998; right now it almost requires a Corporation to coast the development). Also it normalizes the idea of having AI in development, which is in itself dangerous.

[–] misk@piefed.social 2 points 2 days ago* (last edited 2 days ago)

They list iTerm2 as affected but list Linux-specific terminal emulators only as replacement even if there are plenty of those on MacOS. At this point I think those lists are prepared by LLM boosters too.

[–] fruitycoder@sh.itjust.works 4 points 2 days ago (1 children)

Because IF it is superuseful tool and you are being paid to dev then you will have to explain why. Like if a framer showed up to a construction site and refused to use power tools

[–] artyom@piefed.social 4 points 2 days ago (1 children)

But its not. This is more like a framer showed up and you told him to go home so the power tools could build a house that looks like the fucking tower of Pisa.

[–] terabyterex@lemmy.world 10 points 2 days ago (1 children)

this is noy how devs are using ai. they use it as a tool..

non devs may be using ai this way and the house falls apart.