1. Scope and Definitions

• Generative AI tools are systems that produce text, images, code, data, or other content in response to prompts (for example ChatGPT, Gemini, Copilot, Claude, DeepSeek, Midjourney, or DALL·E).
• Assistive tools such as spelling and grammar checkers, reference managers (Mendeley, Zotero), and translation dictionaries that only correct or format text written by the authors are not covered by the disclosure requirement.
• The policy applies to all article types and to every stage: writing, data analysis, figures, peer review, and editorial handling.

2. Authorship

• AI tools do not meet the criteria for authorship: they cannot take responsibility for the work, approve the final version, declare conflicts of interest, or hold copyright. They must not be listed as authors or co-authors.
• AI tools must not be cited as the source of facts, ideas, or findings. Statements taken from AI output must be verified against, and cited to, the original scholarly sources.
• All authors are accountable for the accuracy, originality, and integrity of the entire manuscript, including any part produced with the help of AI.

3. Acceptable Use by Authors

With careful human oversight, authors may use generative AI to:

• improve the language, grammar, clarity, and readability of text they have written themselves;
• assist with translation of their own text, provided the authors check the result for accuracy and terminology;
• brainstorm the structure of a manuscript or search for literature, provided every source is read and verified by the authors;
• help write or debug computer code used in the analysis, provided the code is checked, tested, and described in the methods.

4. Unacceptable Use by Authors

• Generating, fabricating, or manipulating research data, results, statistics, interview transcripts, or field observations.
• Producing the substantive scholarly content of the manuscript, such as the analysis, interpretation, discussion, or conclusions, without critical intellectual contribution and review by the authors.
• Creating or altering images, figures, tables, or photographs reported as research evidence. Adjustments of brightness or contrast are allowed only when they apply to the whole image and do not obscure or add information. The only exception is research in which AI-generated images are themselves the object of study; this must be described in the methods.
• Generating references, or citing sources that the authors have not read. Fabricated or non-existent references are treated as research misconduct.
• Paraphrasing or rewriting existing texts to disguise plagiarism or to evade similarity checking.
• Entering confidential, personal, or sensitive data (for example, data of research participants or community partners) into AI tools that do not guarantee privacy and data protection.

5. Disclosure Statement

If generative AI was used for any purpose other than those of the assistive tools in Section 1, the authors must add a short section entitled Declaration of Generative AI Use before the References, stating the tool (name and version), the purpose, and the parts of the manuscript concerned. Example:

If no generative AI was used, no statement is required. When AI is part of the research method (for example, AI-based data analysis or classification), it must also be described in detail in the Methods section so that the work can be reproduced. The disclosure is published together with the article.

6. Peer Reviewers

• Manuscripts under review are confidential. Reviewers must not upload a manuscript, or any part of it, its figures, or the review report to generative AI tools, because this may breach confidentiality and the authors' rights.
• The scientific evaluation must be the reviewer's own work. Reviewers may use assistive tools to correct the language of the report they have written, as long as no manuscript content is shared.
• Reviewers who suspect undisclosed or inappropriate use of AI in a manuscript should inform the editor rather than make accusations in the report.

7. Editors

• Editors keep submissions confidential and do not upload them to generative AI tools. Editorial decisions are made by people and are never delegated to AI.
• The journal may use established screening tools (for example, similarity checking) as part of its workflow. Results of any AI-detection tool are treated only as an indication, never as proof, and are always assessed by an editor.
• Editors who use AI tools for minor tasks, such as checking the language of their own correspondence, do so without sharing manuscript content.

8. Handling Breaches

Suspected undisclosed or inappropriate use of AI is handled according to the COPE flowcharts. Depending on the stage and seriousness, the editor may ask the authors for clarification and a disclosure statement, request revision, or reject the manuscript. For published articles, the journal may publish a correction, an expression of concern, or a retraction, and may inform the authors' institution in serious cases.

9. Review of this Policy

AI technology and community standards are evolving quickly. This policy is reviewed regularly and may be updated; the version published on this page applies to all manuscripts submitted after its publication. Last updated: October 2026.