AI models ChatGPT and Grok have demonstrated superior performance on the US Medical Licensing Exam compared to the average doctor. ChatGPT scored 98%, significantly higher than the average doctor's score of 75%, while Grok scored 84%. Former FDA chief Scott Gottlieb revealed that ChatGPT not only scored the highest among five top AI models tested but also explained answers like a real doctor. This development raises questions about the potential for AI chatbots to mimic doctors in treatment settings.