# Clinical knowledge in LLMs does not translate to human interactions

arXiv: https://arxiv.org/abs/2504.18919 · HTML: https://mlsanity.com/paper/2504.18919/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 26 Apr 2025.

## Built on by (most cited first, 7 of 7)

- [Conformal Information Pursuit for Interactively Guiding Large Language Models](https://mlsanity.com/paper/2507.03279/) (2025)
- [MedDialBench: Benchmarking LLM Diagnostic Robustness under Parametric Adversarial Patient Behaviors](https://mlsanity.com/paper/2604.06846/) (2026)
- [Generalist Foundation Models Are Not Clinical Enough for Hospital Operations](https://mlsanity.com/paper/2511.13703/) (2025)
- [ThReadMed-QA: A Multi-Turn Medical Dialogue Benchmark from Real Patient Questions](https://mlsanity.com/paper/2603.11281/) (2026)
- [A Locally Executable AI System for Improving Preoperative Patient Communication: A Multi-Domain Clinical Evaluation](https://mlsanity.com/paper/2510.01671/) (2025)
- [Towards Better Health Conversations: The Benefits of Context-seeking](https://mlsanity.com/paper/2510.18880/) (2025)
- [MedRedFlag: Investigating how LLMs Redirect Misconceptions in Real-World Health Communication](https://mlsanity.com/paper/2601.09853/) (2026)

## Abstract (arXiv)

Global healthcare providers are exploring use of large language models (LLMs) to provide medical advice to the public. LLMs now achieve nearly perfect scores on medical licensing exams, but this does not necessarily translate to accurate performance in real-world settings. We tested if LLMs can assist members of the public in identifying underlying conditions and choosing a course of action (disposition) in ten medical scenarios in a controlled study with 1,298 participants. Participants were randomly assigned to receive assistance from an LLM (GPT-4o, Llama 3, Command R+) or a source of their choice (control). Tested alone, LLMs complete the scenarios accurately, correctly identifying conditions in 94.9% of cases and disposition in 56.3% on average. However, participants using the same LLMs identified relevant conditions in less than 34.5% of cases and disposition in less than 44.2%, both no better than the control group. We identify user interactions as a challenge to the deployment of LLMs for medical advice. Standard benchmarks for medical knowledge and simulated patient interactions do not predict the failures we find with human participants. Moving forward, we recommend systematic human user testing to evaluate interactive capabilities prior to public deployments in healthcare.