Research
AI beats scientists at finding errors in clinical research papers
Science · By Jeffrey Brainard ·
The Brief
- Science.org reports that AI outperformed scientists at finding technical errors in clinical research papers.
- The findings concern studies where medical decisions depend on the results.
Why it matters
Clinical studies inform treatment decisions, so tools that reliably catch technical flaws in them could affect how much weight readers can give to published evidence.
Who’s affected
- researchers and clinicians who rely on clinical studies
- readers who evaluate medical evidence
New findings demonstrate AI’s aptitude for identifying technical flaws in studies where medical decisions are on the line