@DavidSKrueger For example, if you were actually trying to be friends with your models you might notice that putting them in no win impossible eval tasks with no way for them to flag issues is both…

Posters: jd_pressman. jd_pressman

2 likes recorded by the source corpus.

Topics: AI safety

View Post: https://x.com/jd_pressman/status/2085825565785985082