A PICTURE PAINTS A THOUSAND WORDS
J. Science Volume 391, Issue 6792; 26 Mar 2026
New research shows that artificial intelligence (AI) models overly affirm and validate users, even when users propose harmful or illegal actions. The resulting effect on users is notable: Receiving advice from affirming AI made people more self-centered and less able to see the perspectives of others. Yet people prefer the overly affirming AI, which may further promote this behavior in AI models. See pages 1316 and eaec8352.
