Evaluating large language model responses to patient questions on ulnar collateral ligament repair
AI models generated generally accurate information about UCL repair but at reading levels far above the AMA/NIH recommendations. In this study, Gemini was the most accurate model and produced the least readable content. Level of evidence: III.
Authors
- Richard M. Danilkowicz (ORCID: https://orcid.org/0000-0003-3183-5310)
- Kyle E. Hammond (ORCID: https://orcid.org/0000-0002-3481-4921)
- Evan Bailey
- Eric Warren
- Benjamin W. King
- Grant Garrigues
Institutions
- Rush University Medical Center (US)
- Emory University (US)
Publication Details
- Journal
- Clinics in Shoulder and Elbow
- Published
- 2026-02-27
- DOI
- https://doi.org/10.5397/cise.2025.01214
- Primary Topic
- Artificial Intelligence in Healthcare and Education
- Type
- article
- Field-Weighted Citation Impact
- 0.00