points by rjsw 1 day ago | 0 comments
The reason for NetBSD rejecting any use of AI is the concern that a future legal case will rule that the output of LLMs trained on GPL code is itself subject to the GPL due to the viral clause in the licence.
[–] F3nd0 1 day ago
I assume the reason was more generally non-permissive licensing and legal issues? Specifically seeing only copyleft as the problem sounds unreasonably sociopathic, so I assume you picked that out as just one prominent example. (That, or I should take care to steer clear of NetBSD.)
[–] spauldo 1 day ago
In context it seems they're concerned about clear ownership of the code. It's lumped in with rules about not submitting code that might not have clear licensing or ownership. Here's the specific part from NetBSD's Commit Guidelines:
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.
[–] ForHackernews 1 day ago
[–] JSR_FDED 1 day ago
[–] NoidFonsense 19 hours ago
Now it's the case that almost all software can be reverse engineered by AI. So the moat will be eroded regardless. Filthy copyleft or not.
[–] voakbasda 16 hours ago
[–] red-iron-pine 18 hours ago
but all the code would be free.
[–] ForHackernews 1 day ago
With the arrival of AI, many of us will need to career change anyway.
[–] throwaway27448 1 day ago
[–] rjsw 22 hours ago
If it is established that copyright is retained then someone else could follow up on the GPL aspects.
[–] lproven 22 hours ago
[–] rjsw 1 day ago
[–] promano 1 day ago
[–] limagnolia 1 day ago
[–] MrMorden 1 day ago