A few days after Micro$lop announced they’re using AI to develop Windows, PCs with Win11 started to have a bug, where the screen goes black (usually no signal), then the GUI slowly crawls back to be visible one-by-one. I wonder what kind of Linux equivalent we will have.
Yeah AI in open source is probably the least negative thing about AI can’t count how many tickets are still open and they are not even that hard to deal with AI seems suitable for the job imo
Probably less Linus outbursts overall since he will have less humanSlop code to review…
Yeah, because the AI vendors sponsoring the project will pull their money out of Linux and create a hard fork that has all the C code converted to Rust with AI.
I think Linus is converting a lot of the kernel to Rust himself tbh.
Linus does very little programming anymore
Im aware, but he is actively using Claude to write some new code for the kernel.
Switch to FreeBSD.
Too popular, it’ll get AI slop eventually too.
We’re gonna have to gird our loins for TempleOS
TempleOS generates random words with a fraction of the energy use of LLMs. The feature is called godspeak.
I love this “A is imperfect so switch your entire existence to B” kind of answer that provides no value.
lol it wasn’t meant to be serious advice. I am not a fan of any of the BSDs on the desktop. I am a happy Fedora user. Linus and the kernel team made their stance on AI clear. We aren’t going to change that. If it gets to the point where it is worse than the alternatives, as a result of AI, I will look elsewhere. Until then, it works.
So you’ll stick around until you’re (possibly the first person you’ve heard about) affected by a (potentially serious, as in data loss) bug?
Maybe! There have always been bugs. I didn’t jump ship any of those times. What matters most is how it compares to the alternatives. What are your plans?
Oh boy. NetBSD! GNU Hurd!
sorry, all the bsd’s are on the list and HURD might actually need it to get anywhere…
2. Do not commit tainted code to the repository.
[…]
Code generated by a large language model or similar technology, such as GitHub/Microsoft’s Copilot, OpenAI’s ChatGPT, or Facebook/Meta’s Code Llama, is presumed to be tainted code, and must not be committed without prior written approval by core.As in, not your code, disavowed.
which one is that?
NetBSD. They are very explicit LLMs are tainted code, and will not be included.
ah, looks like they haven’t looked too deeply at that one since it’s marked “help wanted”. i think their criteria is stricter than just “is there ai code in it”, they’re looking at things like code review too.
