Microsoft AI Diagnoses Better Than Doctors — SkimNews
Get the Tech newsletter
Daily tech — startups, AI labs, chips, the launches that shape the next decade. Free.
- Microsoft developed an AI-powered Diagnostic Orchestrator that achieved 83% accuracy on complex medical cases, significantly outperforming generalist physicians who scored 36% with references and 22% without.
- Microsoft AI published findings in Science showing its system outperformed human doctors on emergency department cases from Beth Israel Deaconess Medical Center, scoring 80% accuracy versus an average of 47% for physicians.
- Microsoft’s Diagnostic Orchestrator uses multiple AI agents powered by OpenAI’s GPT-5 to simulate expert medical reasoning, including test ordering and differential diagnosis, and it did better than standalone frontier AI models.
- Microsoft created SDBench, a new diagnostic benchmark based on real medical case reports, which it plans to open-source to support standardized evaluation of AI diagnostic tools.
- Dr. Dominic King, Microsoft’s AI health lead and study coauthor, says the company plans real-world testing at Beth Israel Deaconess Medical Center within the next few months to assess clinical performance.
- Mustafa Suleyman, Microsoft AI CEO, states the system is a milestone toward 'medical superintelligence' but emphasizes human clinicians will remain central to patient care despite AI advances.
Why it matters: Microsoft’s AI outperformed doctors by 36–58 percentage points across diagnostic benchmarks, suggesting AI could reduce diagnostic errors and lower care costs. If validated in live settings, hospitals and health systems may accelerate AI adoption, shifting clinician roles toward patient interaction rather than data analysis.
Ask SkimNews
