Frequent AI chatbot users more likely to believe anti-vaccine myths, poll finds
TL;DR
A KFF poll of 2,480 US adults found that frequent users of AI chatbots for health advice are more likely to believe vaccine myths. The false claims include vaccines causing autism and the idea that the measles vaccine is more dangerous than measles itself. The correlation remained after accounting for age, education, race and political partisanship, according to the survey. Important caveat: the poll shows correlation, not causation. It does not prove chatbots are directly creating anti-vaccine beliefs.
Nauti's Take
The important question is not only whether chatbots produce false answers. It is also who asks them, and in what state of doubt.
Someone looking for confirmation may get a polite machine instead of a firm boundary. For health AI, source notes and disclaimers are not enough.
Systems need stricter handling of vaccine myths, or they risk turning uncertainty into a scalable service.
Briefingshow
Health advice is a high-stakes test for AI assistants because bad answers can shape real medical decisions. If heavy chatbot use is linked with vaccine myths, a small disclaimer is not enough. Platforms need to show how they verify medical responses, rank sources and prevent dangerous misinformation from being reinforced.
Sources