医学研究人机融合内容透明披露规范与实施指南(2026年版)OA

The rapid iteration of artificial intelligence(AI)tools is propelling medical research into a new phase characterized by human-AI collaboration.In the course of this process,AI has evolved from an exter-nal instrument into a functional component embedded within research workflows.As the collaborative boundary be-tween human researchers and AI becomes increasingly ambiguous,governance challenges have grown increasingly salient.Concurrently,there remains a lack of internationally harmonized standards for the declaration of AI-assis-ted content,giving rise to notable disclosure gaps and implementation disconnects both domestically and abroad.To address these issues,the Professional Committee of Medical Journals of the Society of China University Jour-nals assembled a multidisciplinary expert team covering fields such as medicine,biostatistics,epidemiology,medical ethics,scientific journal editing,and medical AI technology.Employing a modified Delphi method,the team conducted three rounds of online surveys and two consensus meetings.Drawing upon evidence from policy surveys of 1 692 medical journals,position statements from international organizations such as the International Committee of Medical Journal Editors,AI policy texts from the world's major academic publishers(Elsevier,Springer Nature,Wiley,Taylor&Francis,SAGE,etc.),and comparative studies on AI detection tool per-formance,the team ultimately formulated ten recommendations addressing core issues in medical publishing.This guidance emphasizes the core principles of transparent disclosure,human accountability,risk stratification,and technology-assisted oversight,aiming to fully harness the benefits of AI technology while upholding the bottom line of academic integrity.It provides a systematic framework for the standardized disclosure of human-AI collabo-ration content in medical research and offers actionable normative guidance for researchers,journal editors,peer reviewers,and academic institutions.