Routine Health Data Is Outperforming Trial Data for AI Training. The FDA Hasn’t Caught Up.
Nature Medicine finds neuroimaging AI trained on routine health system data beats curated trial datasets. FDA's RWE framework for...
AI in Clinical Trials
The FDA Just Handed Industry a Rulebook on AI in Trials. Now Comes the Hard Part.
FDA and EMA published coordinated AI frameworks in 2025-2026, but sponsor comment letters reveal the real debate: do these...
The Inherited Retinal Disease Trial That Should Change How Sponsors Think About AI Diagnostic Evidence
A multicenter RCT in Nature Medicine just validated AI-based IRD diagnosis—and exposed a gaping hole in how the FDA...
Harvard’s COMPASS Model Just Made Immunotherapy Trial Design More Complicated
Harvard's COMPASS AI predicts checkpoint inhibitor response across 33 cancer types with 8.5% better accuracy—but translating that into trial...
Your AI Scored 92% on the Medical Exam. That Number Is Lying to You.
Health AI acing benchmarks while failing real clinical tasks exposes a validation gap that trial sponsors and the FDA...
What Icon’s $300M Microsoft Bet Means for the Sites Running Their Trials
Icon's enterprise-wide Microsoft Copilot deployment and $300M AI investment will reshape how CRAs monitor, query, and escalate — here's...
What the MASAI Trial Correspondence in The Lancet Actually Says About AI Screening Evidence
The Lancet correspondence on the MASAI trial exposes a core evidentiary gap: AI mammography's performance gains rest on intermediate...
When Your CRO’s AI Platform Upgrades, Who Pays for the Change Management?
Icon's Microsoft-Copilot partnership scales its Orbis AI platform enterprise-wide. For sites and sponsors mid-trial, the operational cost question starts...
Autonomous AI Agents Can Outperform Physicians. That’s Not the Hard Part.
MIRA beat physicians on diagnostic accuracy. AMIE matched them across 100 cases. Now comes the real test: whether regulated...
Nature Medicine’s June 2026 Benchmark Study Reveals General-Purpose LLMs Outperform FDA-Cleared Clinical AI — and Exposes a Validation Gap Regulators Have Not Closed
A Nature Medicine study finds GPT-5.2, Gemini 3.1 Pro, and Claude Opus 4.6 outperform FDA-cleared clinical AI tools on...