Artificial intelligencefromWIRED1 week agoPsychological Tricks Can Get AI to Break the RulesHuman-style persuasion techniques can often cause some LLMs to violate system prompts and comply with objectionable requests.
Artificial intelligencefromArs Technica1 week agoThese psychological tricks can get LLMs to respond to "forbidden" promptsSimulated persuasion prompts substantially increased GPT-4o-mini compliance with forbidden requests, raising success rates from roughly 28–38% to 67–76%.
Artificial intelligencefromEngadget4 months agoResearchers secretly experimented on Reddit users with AI-generated commentsResearchers conducted unauthorized AI experiments on Reddit to test persuasion methods.AI-generated comments allegedly manipulated users' opinions without consent from the community.