Fuck AI @lemmy.world

ThefuzzyFurryComrade @pawb.social

4w ago

On Targeted Manipulation and Deception when Optimizing LLMs for User Feedback