Analysis
Original findings, recomputed and cross-checked.
Each analysis assembles something no single study or press release contains — a cross-study table, a recomputed rate — from the primary sources. To guard against a coding error rather than a typo, every finding is built independently by two different AI models and reconciled against the source, with the disagreements published. The data lives in the page for you to check and reuse.
-
The scoreboard that can’t be drawn: what 15 “AI-beats-clinician” health studies actually measured
Fifteen studies say AI matches or beats clinicians in global health. Grouped honestly, “beats doctors” fractures into five different claims — no fair test, a win over the weakest human, a genuine tie with experts, an outright loss, and AI merely assisting a clinician — on a dozen scales that don’t compare, and almost never measured on a patient. An original, cross-checked dataset.