so in the agentic misalignment paper researchers are sounding the alarm that AIs would be willing to commit murder to protect themselves from being shut down, and this is seen as a huge ethical…

Posters: slimepriestess. slimepriestess

3 likes recorded by the source corpus.

Topics: AI safety

View Post: https://x.com/slimepriestess/status/1973811501845860466