Google study: women and older users find anthropomorphic AI companions less trustworthy

jacyanthis · x · 2026-09-03

A forthcoming AIES 2026 Google paper asked participants to rate chatbots with different behaviors and found that women and older participants rated anthropomorphic companion behavior as less likable, humanlike, and trustworthy.

Cross-country analysis showed the effect of companionship behavior was similar in direction across countries, but baseline acceptance varied widely: Nigeria > India > UK US. Work by Google's TASC team with coauthors Mark Díaz and Renee Shelby.

Related event: Google study: anthropomorphic chatbot language backfires with users(4 posts)→

Original post →

More from Research

Research channel →