Medical publishing in the era of artificial intelligence

The integration of artificial intelligence (AI) into medical research and publishing represents a transformative shift, offering unprecedented opportunities to streamline workflow efficiency, enhance language clarity, and accelerate scientific dissemination. However, this rapid technological adoption introduces ethical, epistemological, and procedural challenges, including concerns over data fabrication, automated bias, intellectual property infringement, and the erosion of scholarly accountability. A structured analysis of practical applications highlights a clear demarcation between acceptable supplementary uses and strictly prohibited practices. AI tools are unacceptable as co-authors, as authorship demands moral, legal, and scholarly accountability that non-human entities cannot possess. Similarly, generating or modifying clinical images and media using AI remains unacceptable due to the risk of data distortion and misrepresentation. Within the editorial domain, AI systems must not replace peer reviewers or editorial decision-makers. Uploading confidential manuscript content into external AI platforms violates author confidentiality, while automated algorithms lack the contextual judgment and critical reasoning necessary for peer evaluation and final publication decisions. Conversely, AI tools are acceptable for planning the structure of a manuscript, language and grammar corrections, translation, quantitative or qualitative data analysis, and formatting, provided their use is transparently disclosed and critically evaluated by human researchers. Ultimately, the ethical deployment of AI in medical literature hinges on the principle that technology must serve as an auxiliary tool rather than a replacement for human cognition. Human authors, peer reviewers, and journal editors retain full responsibility and accountability for the accuracy, integrity, and scientific validity of published work.