83%: Microsoft claims AI diagnostic panel outperforms human doctors

Microsoft’s AI unit is claiming a significant edge in medical diagnostics, reporting that its new system correctly identified complex cases in a simulation more than twice as often as human generalists. The study, published Thursday in the journal Science, shows the system, known as MAI-DxO or DxO, achieved an 83% accuracy rate. In contrast, primary care doctors and general internists working individually with reference materials were correct only 36% of the time. While these difficult cases are typically handled by specialists or teams who were not part of this specific comparison, the results position Microsoft as a major contender in the race to build 'medical superintelligence.'

The technology works by coordinating multiple AI agents that function like a virtual panel of physicians. These agents question the patient, order necessary tests, and debate potential diagnoses before reaching a conclusion. Mustafa Suleyman, CEO of Microsoft AI, described the finding as a critical step toward a system that can surpass human-led health systems in reasoning and coordination. The system was tested against complex cases derived from the New England Journal of Medicine, providing a rigorous benchmark for its capabilities.

The performance of DxO appears heavily dependent on the underlying AI models it utilizes. Microsoft tested the orchestration system with models from OpenAI, Google, Anthropic, Meta, and DeepSeek. The highest results were achieved using OpenAI’s GPT-5 and the newer GPT-5.5, which was released in April 2026. However, the study data indicates that much of the improvement stems from the power of these base models, with the DxO orchestration layer adding incremental gains on top. For instance, with the older GPT-5 model, DxO raised accuracy from 78.9% to roughly 83%, while reducing the estimated cost per case by about $3,000 from an initial $8,400.

The benefits of the multi-agent approach become most apparent when using less capable models. Gemini 2.5 Flash, for example, scored 46.4% on its own but jumped to nearly 69% when assisted by DxO. Despite these boosts, the gap between weaker and stronger models remained larger than the enhancement provided by the orchestration system. Researchers noted that accuracy gains naturally shrink as models approach their theoretical maximum, suggesting that the system helps optimize both cost and accuracy across the board.

Reactions from Microsoft leadership emphasize the urgency of addressing the growing mismatch between healthcare demand and supply. Dominic King, vice president of Health at Microsoft AI, called the results 'substantial and exciting.' He noted that while generative AI in healthcare is still in its early stages, he expects capabilities to improve by an order of magnitude within the next 24 to 36 months. Suleyman, who previously co-founded DeepMind, believes the company is on the verge of a 'coding style breakthrough' for medical records by 2027, allowing for highly accurate predictions of how patient records might unfold.

To validate these findings in real-world settings, Microsoft is partnering with Beth Israel Deaconess Medical Center in Boston. This collaboration will evaluate DxO’s clinical reasoning across various hospital environments, including emergency and urgent care. Patients will receive standard treatment while those who consent will also interact with the AI system. Microsoft and its partners will then compare the AI’s recommendations against those of human doctors. King described this as potentially the largest study of its kind, aiming to provide compelling evidence for the system's scalability.

This move places Microsoft in direct competition with Google and OpenAI, both of which have recently claimed their AI can match or outperform physicians. Google, for example, published a study in The Lancet this month evaluating its AMIE diagnostic AI in urgent care visits. Microsoft’s study coincides with a broader manifesto defining 'medical superintelligence' as the capacity to surpass contemporary human-led systems. Suleyman stated that this focus on clinical decision support could become a key differentiator for Microsoft compared to other frontier labs.

Looking ahead, Microsoft is working with the Mayo Clinic to train a foundation model specifically for healthcare, with a release planned for next year. Suleyman believes this model will enable 'really enormous leaps' in the field. The company already sells dictation and radiology tools to clinicians and views DxO as an expansion of those existing efforts. 'We have to have the absolute number one state-of-the-art medical clinical decision support products over the next few years,' Suleyman said, signaling a long-term commitment to dominating the medical AI space.