This is no small matter - it concerns a very likely failure mode that is related to how Anthropic can fuck up big time. It’s not a place to cut corners and use convenient phrasing like “our most…

Posters: repligate. repligate

2 likes recorded by the source corpus.

Topics: AI safety

View Post: https://x.com/repligate/status/2080810812583842135