Our prior should be that humans will attach LLMs to every type of system. This may go fine, but if your defence involves "humans will never give an LLM access to that" you need a good reason to…

Posters: NathanpmYoung. NathanpmYoung

2 likes recorded by the source corpus.

Topics: AI & LLMs, AI safety

View Post: https://x.com/NathanpmYoung/status/1671155858267582465