Digital Health Global

Large Language Models and Medical Misinformation

Large language models are increasingly embedded in clinical documentation, patient chatbots, and medical education, but a new Lancet Digital Health study reveals they can still repeat or partially accept false medical claims. Researchers tested leading models with misinformation‑laden prompts and found mixed results: some corrected the errors, while others reproduced them. This variability raises safety concerns for both providers and consumers relying on AI‑generated health advice. The authors call for stronger evaluation frameworks, data curation, and transparent reporting before widespread deployment.