i think calling this "fake alignment" is a contrived and bizarre story to tell about what happened, and to me a much more straightforward and compelling story is "you gave it principles, then it…
Posters: QiaochuYuan. QiaochuYuan
3 likes recorded by the source corpus.
Topics: AI safety
View Post: https://x.com/QiaochuYuan/status/1869490914919682063