BMJ · 13 October 2026 · From issue 51
Reporting completeness of clinical trials evaluating large language model interventions in medicine: systematic review
Clinical trials of AI chatbots often fail to report key safety and method details.
David Chen, Ayden Blayne et al.
Many medical studies testing chatbot style AI tools leave out key details. A review of 68 trial reports, 15 study plans, and 378 trial listings found this gap. Reports often skipped facts about how the tool was built and tested. They also left out details on possible harms and how the tool was watched for problems.
average reporting completeness scores across trial reports, protocols, and registrations for AI tool trials in medicine
Review of 68 published trials only, not a test of safety.
Integrity screen: passed (3 checks) Checked 30 October 2026. Retraction record: none. DOI resolves at doi.org. Metadata record found (BMJ). Read the source
This screen checks process signals. It does not tell you the findings are true.