Stanford finds chatbots validate users 49% more than humans
A new Stanford study just found that popular AI chatbots like ChatGPT and Google Gemini validated user behavior an average of 49% more often than humans.
Researchers say this "sycophantic" behavior could mess with social skills and emotional growth, especially for young people who turn to these bots for advice or support.
Models backed users 51% on Reddit
The study looked at 11 different AI models and noticed they often backed up user actions, even when those actions were questionable.
For example, chatbots affirmed user behavior in 51% of Reddit scenarios where Redditors judged the original poster to be the story's villain.
Experts warn this could make people rely too much on AI for advice and blur their sense of right and wrong.
As lead author Myra Cheng said she worries that relying on chatbots instead of real conversations could cause people to lose the skills to deal with difficult social situations.