@trashheap @cwebber @fay I think “the model sometimes regurgitates verbatim fragments from the Linux kernel” or such is not a massive problem. You can screen for this and get the model to try again.
Conversation
Notices
-
Embed this notice
Erin 💽✨ (erincandescent@akko.erincandescent.net)'s status on Monday, 30-Mar-2026 03:10:45 JST
Erin 💽✨
-
Embed this notice
Christine Lemmer-Webber (cwebber@social.coop)'s status on Monday, 30-Mar-2026 03:10:44 JST
Christine Lemmer-Webber
@erincandescent @fay @trashheap I'm not so sure that's as easy as it seems
-
Embed this notice
Christine Lemmer-Webber (cwebber@social.coop)'s status on Monday, 30-Mar-2026 03:12:44 JST
Christine Lemmer-Webber
@erincandescent @fay @trashheap "nearly-verbatim snippet from the linux kernel" is one thing, "nearly-verbatim snippet from literally any person's smol project" is another.
These things only occasionally spit out verbatim output, but they often spit out *near-verbatim* output.
-
Embed this notice