Fuck AI @lemmy.world

ThefuzzyFurryComrade @pawb.social

2mo ago

On Targeted Manipulation and Deception when Optimizing LLMs for User Feedback