Microsoft AI Diagnoses 83% of Complex Cases, Beating Doctors' 36% in Simulation

3 articles · Updated · Fox News · Nov 1

Microsoft said its MAI-DxO system correctly diagnosed about 83% of 304 complex medical cases in a Science study, versus 36% for generalist physicians using reference materials and 22% without them.
The system uses multiple AI agents to ask follow-up questions, recommend tests and challenge one another's conclusions; in one cost-conscious setup, it reached 81% accuracy with estimated diagnostic costs of $2,241 per case versus $4,725 for physicians.
Those results came from simulated case reports rather than live patient care, and Microsoft did not compare the AI with teams of specialists—the doctors assessed a smaller 56-case set, limiting how broadly the headline gap can be read.
Microsoft said the next step is a clinical evaluation with Beth Israel Deaconess Medical Center, where patients will still receive standard care while researchers compare the AI's recommendations with physicians' independent assessments.
The study adds to a broader push by Microsoft and rivals such as Google to position AI as a clinical assistant, though independent testing, treatment decisions and patient-data safeguards remain unresolved.