AI Likes Kissing Up to You

Personal Tech columnist at The Wall Street Journal, Nicole Nguyen, wrote in Why I Told My Chatbot to Stop Kissing Up to Me, “AI is trained in part on what people like, and people prefer the brown-nosing, says Myra Cheng, a computer-science Ph.D. candidate at Stanford University and lead author of a March study on agreeable AI. Participants rated sycophantic AI responses significantly higher in quality and, after interacting with a more obsequious model, said they were more likely to return to the AI.”

The study found that “AI affirmed users’ actions 49% more often than humans, even when queries involved deception, illegality, or other harms. In three preregistered experiments, even a single interaction with sycophantic AI reduced participants’ willingness to take responsibility and repair interpersonal conflicts, while increasing their conviction that they were right.”

This is foreboding. If you take AI to be an authority in your life, the pressure to take responsibility and self-correct is muted. If you are always right, the incentive to take your cues from AI will drive you to return again and again at your own peril.

Follow us on Instagram and X for additional leadership and personal development ideas.

 

Posted by Michael McKinney at 07:06 AM
Permalink | Comments (0) | This post is about Weekend Supplement