Microsoft says MAI-DxO beats generalist doctors on complex cases

Microsoft says its AI diagnostic system, MAI-DxO, correctly diagnosed complex medical cases in a simulation more than twice as often as generalist physicians, according to a study published in Science. The system coordinates several AI agents that question a patient, order tests and discuss a diagnosis.

Using complex cases from the New England Journal of Medicine, MAI-DxO was correct 83% of the time. Primary care doctors and general internists working individually with reference materials scored 36%. The comparison did not include specialists or physician teams, who usually handle cases this complex.

The study also suggests much of the improvement came from the underlying AI models, with Microsoft’s orchestration system adding further gains. With GPT-5, DxO raised accuracy by about four percentage points and cut the estimated cost by about $3,000 from roughly $8,400 per complex case. With GPT-5.5, it raised accuracy on complex cases from 72% to nearly 87% at a slightly lower cost. The gains did not appear on emergency department cases, where GPT-5.5 scored about 85% with or without DxO.

Microsoft AI CEO Mustafa Suleyman called the results a step toward “medical superintelligence.” Microsoft plans to work with Mayo Clinic on a healthcare foundation model, with release planned for next year. It also plans to evaluate DxO with Beth Israel Deaconess Medical Center in Boston, comparing its recommendations with doctors’ for patients who consent while they receive standard care. These are planned evaluations, not reported clinical results.