News for hackers. Stuff that matters.

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

Re:GPT-6 Astra covertly underperforms on safety research by erincandescent on Thursday November 05, 2026 @11:28AM (#50381125)
Because the CoT is supposed to be the independent control. If you RL on the CoT, then you lose it as a means of understanding the internal workings of the mdoel
[ Parent | on HN ]

There are no comments on this comment yet.