News for hackers. Stuff that matters.
Re:Tell HN: Received a CV rigged with an injected LLM prompt by JohnFen on Friday November 01, 2024 @10:03AM (#42016993)
> you can't tell the injected statement is actually false.

It's dishonest on its face because it's trying to subvert the theoretical automation in order to get it to produce a false report. That's overt lying in my view, but even if you don't consider is so, it is at least lying-adjacent and is entirely dishonest.

There are no comments on this comment yet.