Artificial Intelligence (AI) Policy

Last reviewed August 2026 · Applies to all submissions to JMVRA (ISSN 2984-8911)

Large language models and other generative tools are now part of how research is written. JMVRA does not prohibit them. What it requires is that their use is declared, and that a human being remains answerable for every sentence, number and citation in the published paper. An AI tool cannot take responsibility for a claim, cannot consent to publication and cannot handle a post-publication query — which is precisely why it cannot be an author.

Permitted — With Disclosure

How to Declare AI Use

Add a short statement before the reference list. Name the tool, its version if known, and what it was used for. Two sentences is enough:

• Example — language editing: “The authors used [tool name] to improve the grammar and readability of the manuscript. All content was written, reviewed and verified by the authors, who take full responsibility for it.”
• Example — code assistance: “[Tool name] was used to assist with writing the analysis scripts. The study design, data and interpretation are entirely the authors’ own, and all output was independently validated.”
• Example — no AI used: No statement is required. Silence is read as “no generative tools were used”, which is why an undeclared use is treated as a false declaration.

Fabricated Citations — Read This

The most common problem we see is not plagiarism. It is citations that do not exist. Language models generate references that look correct — plausible authors, a real journal, a believable title and page range — for papers that were never written. Reviewers check references, and a manuscript containing invented sources is rejected regardless of how good the rest of it is.

Reviewers and Editors

• Manuscripts are confidential. A reviewer may not paste any part of a manuscript under review into a public AI tool. Doing so uploads unpublished work to a third party and breaches the confidentiality undertaking in the Peer Review Policy.
• The review itself must be the reviewer’s own judgement. AI may polish the wording of a report; it may not form the assessment.
• Editorial decisions are made by people. No accept, reject or revise decision at JMVRA is made by an automated system.
• Screening tools assist, they do not decide. Similarity and AI-detection scores are read by an editor alongside the manuscript, never applied as an automatic threshold.