News for hackers. Stuff that matters.

A comment on LLMs are too easy to automatically red team into toxicity. Parent comment

Re:LLMs are too easy to automatically red team into toxicity by sokoloff on Monday July 03, 2023 @06:37PM (#36579725)
I don’t like it, but the usage is “[attack as a] red team”, in other words “act in an adversarial manner”.
[ Parent | on HN ]

There are no comments on this comment yet.