AI in diagnosis, drug discovery, hospital operations, and medical devices. We separate a preprint, a peer-reviewed result, a regulatory authorization, and evidence that patients actually benefit.
Start here
ExplainerMedicineNew
A medical AI result can be a preprint, a peer-reviewed study, a regulatory authorization, or proof that patients benefit, and each one means something different.
Evidence: 2 sources · 2 official records
NewsAI-prepared · review pendingMedicineNew
AstraZeneca and Oracle plan to test AI in clinical research, starting with hypertension care in Abu Dhabi; the announcement does not establish patient benefit.
Evidence: 2 sources · 1 official record · 1 independent report · 2 open questions
NewsAI-prepared · review pendingMedicineNew
Mass General Brigham announced FAHR-Face research published in The Lancet Digital Health; survival associations do not show that using the tools improves treatment.
Evidence: 3 sources · 1 official record · 1 independent report · 1 research paper · 2 open questions
NewsMedicineNew
Public comments on the FDA's discussion paper about regulating generative AI medical devices closed on 19 October, with hospitals asking for clearer risk and labeling rules; the paper is not guidance and binds no one.
Evidence: 5 sources · 3 official records · 2 independent reports · 3 open questions
Research watchMedicineNew
A peer-reviewed Lancet study found Google's AMIE diagnostic chatbot could take pre-visit histories from 98 patients without a safety stop, but it was a small single-site study with no control group and did not measure patient outcomes.
Evidence: 5 sources · 2 official records · 1 independent report · 2 research papers · 4 open questions
NewsMedicineNew
The UK government accepted all 44 recommendations of the MHRA-convened National Commission on regulating AI in healthcare, including lifelong monitoring of AI devices, but the detailed implementation plan is not due until spring 2027.
Evidence: 4 sources · 2 official records · 2 independent reports · 3 open questions
