Microsoft’s AI diagnostic system, MAI-DxO, correctly diagnosed 83% of complex cases drawn from The New England Journal of Medicine in a simulation, against 36% for generalist physicians using reference materials, according to a study published Thursday in Science, TechCrunch (opens in new tab) reported. Physicians without reference tools scored 22.3%, Medical Economics (opens in new tab) reported.

On emergency-department cases from the MIMIC-IV-ED database, the system scored 80% against 47% for physicians, Microsoft and Medical Economics (opens in new tab) reported. The system coordinates several AI agents that act like a panel of doctors, asking questions, ordering tests and debating a diagnosis.

TechCrunch (opens in new tab) noted that cases this complex are usually handled by specialists, who were not part of the comparison, and that the study’s data suggests much of the gain comes from stronger underlying models. On emergency cases, GPT-5.5 scored about 85% with or without the orchestration system, TechCrunch (opens in new tab) reported.

“It needs clinical validation,” Microsoft AI (opens in new tab) CEO Mustafa Suleyman told The Verge (opens in new tab). “That is a leap, but this is a major milestone along that journey.” The system has only been tested on past cases, The Verge (opens in new tab) reported.