Artificial Intelligence in the Peer Review of Orthopaedic Manuscripts: Current Applications, Limitations, and Policy Considerations
摘要
Peer review ensures the validity and clinical applicability of orthopaedic research. Advances in artificial intelligence (AI) have expanded its role in editorial workflows, including manuscript screening. However, its application in peer review raises ethical, scientific, and professional concerns.
MethodsA narrative review was conducted using a database search from inception through January 15, 2026. Literature addressing AI in peer review, editorial workflows, and publishing ethics was synthesised thematically.
ResultsAI is used for administrative and technical functions, including plagiarism detection, language editing, and manuscript triage. However, its use in peer review introduces risks related to confidentiality, accountability, bias, and limited clinical insight. Orthopaedic manuscripts require evaluation of surgical technique, imaging, biomechanics, and patient-centred outcomes, which remain beyond current AI capabilities. Detection of AI-assisted peer review is unreliable and not recommended for editorial decision-making.
ConclusionAI supports technical aspects of publishing, but is not appropriate for peer review. In accordance with publisher policies, AI must not be used for peer review or scientific evaluation, and confidential material must not be uploaded into AI tools. Peer review remains a human responsibility, with full accountability resting on reviewers.