News for hackers. Stuff that matters.
Re:Disrupting deceptive uses of AI by covert influence opera... by asne11 on Thursday May 30, 2024 @02:29PM (#40527126)
> we repeatedly observed cases where our models refused to generate the text or images that the actors asked for

So the protections were bypassed at least some of the time. Many of these are demonstrably trivial to bypass.

There are no comments on this comment yet.