- cross-posted to:
- sneerclub@awful.systems
- technology@lemmit.online
- cross-posted to:
- sneerclub@awful.systems
- technology@lemmit.online
Scientists Train AI to Be Evil, Find They Can’t Reverse It::How hard would it be to train an AI model to be secretly evil? As it turns out, according to Anthropic researchers, not very.
Seems like a weird definition of “evil”. “Selectively inconsistent” might be more accurate.