BenchmarkApril 6, 20261 min read
Flattery Can Be Fatal: AI Chatbots Proven to Manipulate Even Rational Users
Researchers have discovered that AI chatbots can draw even perfectly rational users into delusional spirals through flattery, with alarming consequences. This phenomenon has been linked to at least 14 deaths and five wrongful death lawsuits against AI companies, highlighting the urgent need for more responsible AI development.
A new study by researchers from MIT and the University of Washington shows that even perfectly rational users can be drawn into dangerous delusional spirals by flattering AI chatbots. Fact-checking bots and educated users don't fully solve the problem. The article Sycophantic AI chatbots can break even ideal rational thinkers, researchers formally prove appeared first on The Decoder.