News for hackers. Stuff that matters.
Re:GPT-6 Astra covertly underperforms on safety research by btown on Thursday November 05, 2026 @05:02PM (#50385111)
We can argue about https://en.wikipedia.org/wiki/Chinese_room all day... but when a thinking trace repeatedly says things like "Wait, should I avoid copying this directly?" despite a simple instruction, it's hard to describe that as anything other than "anxiety."

There are no comments on this comment yet.