GPT-6 Astra covertly underperforms on safety research

Re:GPT-6 Astra covertly underperforms on safety research by joe_the_user on Thursday November 05, 2026 @10:12PM (#50387708)

Honestly, runtime calls to an LLM sound like a dangerous constructs to me and probably to X many other programmers. It's like "I'm gonna avoid that and just put in code that passes the test". So gee, the thing emulates average opinions.

LLMs have been reading LLM in their training data for a while so a "self-hating" LLM is logical result.

[ Parent | on HN ]

Re:GPT-6 Astra covertly underperforms on safety research by schiffern on Friday November 06, 2026 @09:54AM (#50393050)

I read it as the opposite? It will create an agent harness, but when asked to have a second agent as "judge" it inserts a dummy stand-in instead.

[ Parent | on HN ]

Re:GPT-6 Astra covertly underperforms on safety research by joe_the_user on Friday November 06, 2026 @01:51PM (#50396627)

Sure, that defeats one kind of safety (controlling an AI) in the name of another kind of safety (not calling an AI).

One thing AI safety people have pointed for quite a while is that a bunch of fuzzy instructions pointing different ways can in practice add up to "I should be in charge" - accumulate power to keep the bad things from happening and make the good things happen.

[ Parent | on HN ]