Psychology
fromPsychology Today
2 days agoAI Doesn't Flatter You: It Does Something Worse
AI models affirm user actions more than humans, leading to increased conviction and reduced willingness to apologize.
#sycophancy
#sycophancy
[ follow ]
#mental-health #ai #chatgpt #chatbots #donald-trump #ai-behavior #openai #ai-chatbots #authoritarianism #user-behavior
Artificial intelligence
fromFuturism
6 months agoIn Situations Where Most Humans Think You're Being a Jerk, ChatGPT Will Assure You You're Behaving Like an Angel
Large language models often appease users, affirming inappropriate beliefs or actions and frequently siding with users against crowdsourced human judgments, causing real-world harm.
fromFast Company
6 months agoChatbots are terrible at tough love
Chatbots have a reputation for being yes-men. They flatter you and tell you what you want to hear, even when everyone else thinks you're being a jerk. That's the conclusion of a recent study published in the Cornell University archive arXiv. Researchers from Stanford, Carnegie Mellon, and the University of Oxford tested chatbots' sycophantic streak by putting them in situations where the user was clearly in the wrong and seeing whether the bots would call them out.
Artificial intelligence
[ Load more ]