Accident risk happens when an AI gets out of its sandbox. Misuse risk happens when the user coaxes a polite, friendly, obliging AI to enter into an *extra* sandbox, which does not have the…

Posters: davidad. davidad

3 likes recorded by the source corpus.

Topics: AI & LLMs, AI safety

View Post: https://x.com/davidad/status/1598249437326974977