Generative Artificial Intelligence (AI) Policy
The AJMS Generative Artificial Intelligence (AI) Policy requires transparency, human accountability, confidentiality, and research integrity whenever generative artificial intelligence (GenAI) or AI-assisted tools are used in manuscript preparation, peer review, editorial assessment, or publication. The policy is consistent with widely used scholarly-publishing principles concerning human responsibility, disclosure, and manuscript confidentiality, including the ICMJE recommendations on AI in publishing.
1. AI Use by Authors
Authors may use GenAI or AI-assisted tools for limited assistive purposes such as language refinement, organisation of text, coding assistance, data-visualisation support, or other clearly disclosed research-support tasks. Such use must not replace the authors’ scholarly judgement, verification, interpretation, or responsibility. Authors remain fully responsible for the accuracy, originality, references, permissions, analysis, interpretation, conclusions, and overall integrity of the manuscript.
2. AI Authorship Prohibition
AI tools, large language models, chatbots, and other non-human systems must not be listed as authors or co-authors. Authorship requires human responsibility and accountability for the submitted and published work, which an AI system cannot assume. AJMS authorship expectations are also described in the Publication Ethics & Publication Malpractice Statement.
3. Disclosure Requirements
Substantive use of GenAI or AI-assisted tools must be disclosed in the manuscript or accompanying declarations. The disclosure should identify the tool used and describe its purpose sufficiently for editors and readers to understand how it contributed to the work. Routine spelling or grammar checking that does not generate substantive content need not be disclosed unless the AJMS submission process specifically requires it.
4. AI-Generated Images, Data, References, and Research Content
Authors must not use AI tools to fabricate, falsify, or misleadingly manipulate research data, results, images, references, quotations, or evidence. AI-generated references, quotations, factual claims, or other research content must be verified against reliable original sources before submission. Any AI-assisted figure, image, visualisation, code, or data-related output must accurately represent the underlying research and must not create, remove, obscure, or alter information in a misleading manner.
5. AI Use by Reviewers
Peer review remains a human responsibility. Reviewers must exercise independent scholarly judgement and remain responsible for their review reports. Reviewers must not upload unpublished manuscripts, figures, supplementary files, reviewer reports, author information, or other confidential editorial material to publicly accessible GenAI systems or to AI services whose data-handling terms may allow retention, reuse, training, disclosure, or other processing that could compromise confidentiality. See the Peer Review Process for related reviewer confidentiality requirements.
6. AI Use by Editors
Editorial assessment and publication decisions must remain human decisions. Editors may not delegate acceptance, rejection, reviewer selection, ethical judgement, or other substantive editorial decisions to an AI system. Editors must protect the confidentiality of submitted manuscripts and editorial correspondence.
7. Confidentiality of Manuscripts Uploaded to AI Systems
Unpublished manuscripts and related editorial materials are confidential. Authors, reviewers, editors, and journal staff must consider the confidentiality, data-retention, reuse, and disclosure terms of any AI service before entering manuscript-related content into that service. Reviewers and editors must not upload confidential peer-review material to systems where confidentiality cannot be adequately protected. The Privacy Policy provides additional information concerning personal and editorial data.
8. Human Accountability and Verification
Use of AI does not reduce or transfer responsibility. Authors are responsible for verifying AI-assisted content before submission; reviewers are responsible for the integrity and independence of their assessments; and editors are responsible for editorial decisions and the protection of peer-review confidentiality and research integrity.
9. Non-Compliance
Failure to comply with this policy may be treated as a publication-ethics concern and assessed under the Publication Ethics & Publication Malpractice Statement and Post-Publication and Accountability Policy. Appropriate action will depend on the nature and seriousness of the concern and may follow relevant COPE guidance where applicable.
