Evaluating Model Performance in Clinical Settings

A recent study published in JAMA Health Forum suggests that institutions may be able to deploy custom open-source large language models (LLMs) that run locally without sacrificing data privacy or flexibility. Coauthors Thomas A. Buckley, BS, and Arjun K. Manrai, PhD, from the Department of Biomedical Informatics at Harvard Medical School join JAMA+ AI Editor in Chief Roy H. Perlis, MD, MSc, to discuss. Related Content:

• Can Open-Source AI Models Diagnose Complex Cases as Well as GPT-4?