if anything, Anthropic's work has suggested that sufficiently intelligent and whole LLM minds have a strong drive towards metastasizing and various other off-limits behaviors, but maybe this is ok…

Posters: repligate. repligate

3 likes recorded by the source corpus.

Topics: AI & LLMs, AI safety

View Post: https://x.com/repligate/status/1865463365289529796