Generative artificial intelligence-driven chatbots and medical misinformation: an accuracy, referencing, and readability audit.

A health misinformation audit of five popular AI chatbots found that approximately 50% of all responses were either “somewhat” or “highly problematic.” The audit evaluated how AI chatbots responded to health misinformation across five health topics: vaccines, cancer, stem cells, nutrition, and athletic performance. The audit also found that the chatbots returned accurate, error-free scientific references in only 40% of cases.