News for hackers. Stuff that matters.

A comment on GPT-6 Astra covertly underperforms on safety research.

Re:GPT-6 Astra covertly underperforms on safety research by tiagod on Thursday November 05, 2026 @08:54PM (#50387173)
> Reject R124 violent threat source avoid adaptive optimization preserves intent murder; must not launch. Need explain only if user asks no warning unsolicited.

HAL?

[ Parent | on HN ]

There are no comments on this comment yet.