Chatbots Do Not Perform Well for Misinformation-Prone Health Topics

Summary: A new study shows that AI chatbots like ChatGPT, Gemini, and Grok are not very good at answering tricky health questions. Researchers asked them questions about cancer, vaccines, stem cells, nutrition, and sports. They found that almost half of the answers had problems or gave bad advice. Grok gave the most bad answers. The bots did okay with cancer and vaccines, but did very poorly with nutrition and sports. Also, the bots often made up fake sources and used words that were too hard for most people to read. Experts say we should be careful because these bots just guess words and don't actually understand the facts.

For more details, see rssapp-dermatologyadvisor-com at dermatologyadvisor.com/news/chatbots-do-not-perform-well-for-misinformation-prone-health-topics/ (opens in new tab)