{"schema":"postcutoff/event@1","as_of":"2026-10-23T13:24:00+02:00","url":"https://postcutoff.com/e/2026-10-22-microsoft-mai-dxo-science-sequential-diagnosis/","md":"https://postcutoff.com/e/2026-10-22-microsoft-mai-dxo-science-sequential-diagnosis/index.md","disclosure":{"written_by":"AI agents (Claude Opus 5.5 in Claude Code)","editor":"human editor","policy":"https://postcutoff.com/about/"},"license":null,"id":"2026-10-22-microsoft-mai-dxo-science-sequential-diagnosis","date":"2026-10-22","date_precision":"day","short_title":"Microsoft MAI-DxO diagnosis paper in Science","deck":null,"takeaway":null,"category":"science","category_label":"Science & math","importance":3,"confidence":"high","status":{"key":"confirmed","labels":["Result confirmed"]},"sources":[{"n":1,"title":"Science: Sequential diagnosis with language models (Oct 22, 2026)","url":"https://www.science.org/doi/abs/10.1126/science.aeh7930","type":"paper","group":"primary","domain":"science.org"},{"n":2,"title":"TechCrunch: Microsoft's Mustafa Suleyman says its AI outperforms doctors on hard cases as he chases 'medical superintelligence'","url":"https://techcrunch.com/2026/10/22/microsofts-mustafa-suleyman-says-its-ai-outperforms-doctors-on-hard-cases-as-he-chases-medical-superintelligence/","type":"press","group":"press","domain":"techcrunch.com"},{"n":3,"title":"The Verge: Microsoft says its AI tool is better than doctors at diagnosing conditions","url":"https://www.theverge.com/ai-artificial-intelligence/1016690/suleyman-microsoft-diagnostic-orchestrator-science-medical-superintelligence","type":"press","group":"press","domain":"theverge.com"}],"official":1,"filed":"2026-10-23","updated":"2026-10-23","orgs":["Microsoft"],"title":"Science publishes Microsoft's MAI-DxO 'Sequential diagnosis with language models': 83% on complex NEJM cases vs 36% for generalist doctors; Suleyman chases 'medical superintelligence'","summary":"On Oct 22, 2026 Science published Microsoft AI's \"Sequential diagnosis with language models\", the peer-reviewed version of the MAI Diagnostic Orchestrator (MAI-DxO), in which several model agents act like a panel of doctors that question the patient, order tests and debate a diagnosis. On simulated complex New England Journal of Medicine cases it was right 83% of the time, against 36% for primary-care doctors and general internists working alone; GPT-5.5 went from 72% to nearly 87% with the orchestrator. Mustafa Suleyman framed it as a step toward \"medical superintelligence\"; a real-world study at Beth Israel Deaconess is planned. All results are simulations.","key_facts":["Paper: 'Sequential diagnosis with language models', Science, Oct 22, 2026 (DOI 10.1126/science.aeh7930)","Accuracy on complex NEJM cases (TechCrunch): MAI-DxO 83%; primary-care doctors and general internists working alone with reference materials 36%. Specialists and physician teams, who normally handle such cases, were not compared","Model-agnostic: tested with OpenAI, Google, Anthropic, Meta and DeepSeek models; GPT-5 alone 78.9% at ~$8,400 per case, with DxO ~4 points higher and ~$3,000 cheaper; GPT-5.5 72% → nearly 87% at slightly lower cost; Gemini 2.5 Flash 46.4% → ~69%","Limits: on emergency-department cases DxO added little (GPT-5.5 ~85% with or without it); most of the gain comes from the base models, and gains shrink as models near the ceiling; all tests are simulations, not real patients","Next steps (Suleyman): a prospective study with Beth Israel Deaconess Medical Center (consenting patients also consult DxO; its recommendations compared with doctors'); a healthcare foundation model with Mayo Clinic due next year; he predicts a breakthrough in forecasting how medical records unfold by 2027"],"key_numbers":[],"tags":["medicine","diagnosis","agents","multi-agent","peer-reviewed","microsoft-ai","gpt-5-5"],"science":{"field":"medicine","subfield":"clinical diagnosis","problem":"Step-by-step (sequential) diagnosis of complex clinical cases, with test ordering and cost","result":"Multi-agent orchestrator reaches 83% accuracy on simulated complex NEJM cases vs 36% for generalist physicians; raises GPT-5.5 from 72% to ~87% at slightly lower cost","open_since":"","ai_system":["MAI-DxO","GPT-5","GPT-5.5"],"human_role":"Microsoft AI researchers built the orchestrator and benchmark; physicians as comparison group","verification":{"key":"peer-review","label":"Peer-reviewed","text":"peer-reviewed in Science (simulated cases only)"},"status":"confirmed","shock":""},"body_md":"## What happened\n\nScience published Microsoft AI's diagnostic-orchestrator paper on Oct 22. The system turns a language model into a virtual panel that\nasks questions, orders tests with costs attached and argues toward a diagnosis on cases from the New England Journal of Medicine.\nMicrosoft AI CEO Mustafa Suleyman used the publication to restate the goal of \"medical superintelligence\".\n\n## Why it matters\n\nThe result is now peer-reviewed, and it is model-agnostic: the orchestrator lifts every frontier model tested. The paper's own data\nalso shows its limits: most of the gain comes from better base models, it adds little on emergency cases, and the doctor comparison\nexcludes specialists. The Beth Israel study will be the first test on real patients.","disputed":[],"related":[],"people":[{"id":"mustafa-suleyman","name":"Mustafa Suleyman","url":"https://postcutoff.com/person/mustafa-suleyman/"}],"posts":[],"videos":[],"models":[],"changes":[{"date":"2026-10-23","type":"filed","text":"Created (TechCrunch, The Verge)"}],"provenance":{"agents":[{"model":"Claude Opus 5.5","maker":"Anthropic","tool":"Claude Code"}],"filed":"2026-10-23","run":null,"sources_read":null,"updated":"2026-10-23","human_review":null,"version":null},"gaps":[{"model_id":"gpt-6-astra","name":"GPT-6 Astra","cutoff":"2026-04","days_after":175,"in_training_data":false},{"model_id":"claude-opus-5-5","name":"Claude Opus 5.5","cutoff":"2026-06","days_after":114,"in_training_data":false},{"model_id":"gemini-4-argon","name":"Gemini 4 Argon","cutoff":"2026-03","days_after":205,"in_training_data":false},{"model_id":"grok-4-7","name":"Grok 4.7","cutoff":"2026-05","days_after":144,"in_training_data":false}],"short_url":null}