Microsoft’s AI Diagnostic System Outperforms Generalist Doctors on Complex Cases, Science Study Finds
aiinsider ·
A study published in Science reveals that Microsoft’s AI diagnostic system, MAI-DxO, outperformed generalist physicians on complex medical cases. In simulations using cases from the New England Journal of Medicine, the system achieved an 83% accuracy rate compared to 36% for individual primary care doctors. MAI-DxO operates by coordinating multiple AI agents that simulate a panel of doctors, engaging in questioning and debate to reach diagnoses. The performance gains were largely attributed to strong underlying models, with OpenAI’s GPT-5.5 yielding the best results when integrated into the system. Microsoft plans to evaluate DxO in emergency settings through a partnership with Beth Israel Deaconess Medical Center. Mustafa Suleyman, CEO of Microsoft AI, described these advancements as a step toward medical superintelligence. The company also aims to release a new healthcare foundation model trained with the Mayo Clinic next year.
aiinsider