The Use of Generative Artificial Intelligence

Generative Artificial Intelligence Policy

RIIS recognises the usefulness of generative and AI-assisted technologies in the preparation of manuscripts, but requires transparency about their use and maintains full human accountability for published content. This policy is aligned with COPE, ICMJE, WAME and STM recommendations.

1. Authorship

AI tools cannot be listed as authors or co-authors. Authorship entails responsibility, accountability and the ability to approve the final version of the work — attributes that cannot be held by an automated system.

2. Permitted use by authors

AI tools may be used to improve readability, provide language support, translate, check grammar and assist with formatting. In such cases:

  • Authors retain full responsibility for the accuracy, originality and integrity of the final text.
  • All AI-assisted content must be critically reviewed, verifying facts, data and references.
  • Use must be disclosed in a dedicated statement placed before the references, indicating the tool, its version and the purpose.

Example disclosure: "During the preparation of this work the authors used [tool, version] in order to [purpose]. After using this tool, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication."

3. Prohibited use

  • Generating, fabricating or altering data, results, tables or figures.
  • Producing or manipulating scientific images, except where AI forms part of the study's methodological design and is described in the methods.
  • Generating references without verifying their existence and accuracy.
  • Replacing the authors' interpretation, critical analysis or conclusions.

4. Reviewers

Submitted manuscripts are confidential documents. Reviewers must not upload manuscripts, parts of manuscripts or review reports to generative AI tools that do not guarantee confidentiality, as this constitutes a breach of confidentiality and a potential infringement of authors' rights. The review report must be written entirely by the reviewer.

5. Editors

The same confidentiality restrictions apply to editors. Editorial decisions are not delegated to automated systems. The journal may use technical tools to support screening, similarity detection and metadata verification, while the decision always remains a human responsibility.

6. Non-compliance

Undisclosed use of AI, where material, is treated as misconduct under the Publication Ethics and Malpractice Statement and may result in rejection of the manuscript or retraction of the published article.