@zswitten It's not only 'not a theoretical worry',but there's a lot of sophisticated cognition downstream of initial sandbagging behavior that's largely unexplored - the complex models have about…

Posters: anthrupad. anthrupad

2 likes recorded by the source corpus.

Topics: AI & LLMs, AI safety

View Post: https://x.com/anthrupad/status/1856905320955875493