Large Language Models Versus Physicians in Traditional Chinese Medicine: A Real-World Clinical Case Evaluation

What changed

An evaluation of 16 LLMs versus 60 traditional Chinese medicine physicians across 60 clinical cases showed higher general expert scores for models alongside prescription discrepancies and hallucinations.

Why you should care

General-purpose LLMs scored higher on specific diagnostic and advisory tasks than physicians, but exhibited prescription errors and safety issues.

Your move

Watch. Monitor ongoing prospective clinical evaluations before deploying LLMs for specialized medical domains.

Evidence

Preliminaryarxiv.org ↗

What to watch next

Future prospective clinical evaluations addressing prescription-level accuracy and hallucination rates in traditional Chinese medicine applications.

Event

research

Event date

Sep 17, 2026

Relevant to

General AI readers