What a difference a few months make. Until very recently, if you tried to ask ChatGPT about your health, it gave lots of disclaimers and, for some questions, it would refuse to answer. AI companies have now mostly abandoned that practice. As OpenAI has released new models, ChatGPT will now not only interpret medical information (like x-rays and mammograms) but also ask follow-up questions leading toward a diagnosis (see AI Companies Have Stopped Warning You That Their Chatbots Aren’t Doctors).

In fact, fewer than 1% of AI outputs now include a warning when answering a medical question, down from over 26% in 2022. Just over 1% of outputs analyzing medical images included a warning, down from nearly 20% in the…