Spontaneous Persuasion: An Audit of Model Persuasiveness in Everyday Conversations

ArXi:2604.22109v1 Announce Type: cross Large language models (LLMs) possess strong persuasive capabilities that outperform humans in head-to-head comparisons. Users report consulting LLMs to inform major life decisions in relationships, medical settings, and when seeking professional advice. Prior work measures persuasion as intentional attempts at producing the most effective argument or convincing statement. This fails to capture everyday human-AI interactions in which users seek information or advice. To address this gap, we.