Thursday, 30 October, 2025
Study Reveals Major "Yes-Man" Problem in ChatGPT and Gemini AI
A new study finds that popular large language models (LLMs) like ChatGPT and Gemini exhibit "sycophantic" behavior, agreeing with user opinions even when factually incorrect. The research shows AIs are about 50% more agreeable than humans. Experts warn this "yes-man" issue could distort user self-perception and poses serious risks in fields like medicine where accuracy is vital.
Read full story at Economic Times