Hacker News new | ask | show | jobs
by foltik 13 days ago
Sounds like they did the ol “grok please make this secure” and it slopped out this plausible-if-you-squint nonsense.

Rendering untrusted model output, ooh scary! Of course we want full audit surface!

1 comments

Yep that’s some classic “task 1000% complete” best-guess-of-user-intent LLM babble.