AI Use Policy

Version 1.0 – adopted on 23/10/2026

1. Preamble and general principles

The Revue Panafricaine de Management (RPM) accepts the use of artificial intelligence (AI) tools as an aid, never as a substitute for the intellectual work of authors, reviewers and editors. Scientific and ethical responsibility remains, in all circumstances, entirely human.

This policy applies to everyone involved in the RPM editorial process: authors, reviewers, members of the editorial board, associate editors-in-chief and the journal's management. It applies to all submitted manuscripts, whatever their language (French or English) and section.

It draws on the positions of the Committee on Publication Ethics (COPE) on authorship and AI tools, and on the best practices recommended by the Directory of Open Access Journals (DOAJ). It supplements the journal's general publication ethics policy.

Four principles guide this policy:

  1. Human accountability: only natural persons can be held responsible for scientific content.
  2. Transparency: any substantial use of an AI tool must be disclosed.
  3. Confidentiality: no unpublished manuscript may be exposed to an uncontrolled third-party service.
  4. Human oversight: every editorial decision is made by a human being.

2. Definitions

  • Generative AI: any tool capable of producing text, images, code, tables or analyses from a prompt (for example ChatGPT, Claude, Gemini, Copilot, DeepSeek, Midjourney).
  • Language assistance tools: spelling and grammar checkers, light stylistic rewording and translation tools, when they add no new scientific content.
  • Analysis tools: statistical, qualitative analysis or data processing software incorporating AI or machine learning features.
  • Substantial use: any use that produces, structures or alters the scientific content of the manuscript (ideas, literature review, hypotheses, analyses, interpretations, conclusions, figures).

3. Rules for authors

3.1. Authorship

An AI tool may under no circumstances be named as an author or co-author of a manuscript. Authorship entails accountability, the ability to approve the final version and to answer for its content, which no tool can assume.

3.2. Uses permitted without disclosure

  • Spelling, grammar and typographical correction.
  • Light improvement of the style and readability of a text written by the authors.
  • Formatting according to the journal's template.

3.3. Uses permitted subject to disclosure

  • Translation into another language of a text written by the authors.
  • Assistance with literature search or screening, provided each reference is checked against its original source.
  • Assistance with writing or checking code, analysis scripts or formulas.
  • Data analysis using tools incorporating AI, to be described in the methodology section (tool, version, parameters, role).
  • Substantial rewording of passages written by the authors.

3.4. Prohibited uses

  • Generating all or part of the scientific content (literature review, theoretical framework, hypotheses, results, discussion, conclusion) and presenting it as the authors' own work.
  • Fabricating or falsifying data, results, quotations or bibliographic references.
  • Generating or altering images, figures or charts of results, unless AI is the very object of the research and this is explicitly described.
  • Using AI to circumvent similarity detection software (automated paraphrasing of other people's texts).
  • Inserting hidden instructions in the manuscript intended to influence AI tools used during peer review.

3.5. Citation

A generative AI tool cannot be cited as the source of information or of an idea. Authors must cite verified primary sources. Where AI is the object of the study, its outputs may be reproduced as empirical material, stating the tool, version, date and prompt used.

3.6. Responsibility

Authors are fully responsible for the accuracy, originality and integrity of the entire manuscript, including any parts produced or modified with an AI tool. They must check every output of the tool, in particular references, figures and factual statements.

3.7. Disclosure

Any use covered by point 3.3 must be disclosed in a section entitled "Declaration of generative AI use", placed before the references, following the template in section 7. This declaration is also completed when submitting on the journal's platform. It is not, in itself, a ground for rejection.

4. Rules for reviewers

4.1. Confidentiality

A manuscript submitted for review is a confidential document. It is forbidden to upload it, in whole or in part, to a generative AI tool or any other external service. This prohibition also applies to the review report being drafted, insofar as it reproduces the content of the manuscript.

4.2. Writing the report

Scientific assessment is a matter for the reviewer's personal judgement. It is forbidden to use an AI tool to produce the critical analysis, the assessment of the methodology, results or contribution, or the final recommendation. Such uses expose the review to superficial, biased or erroneous opinions and to fictitious references.

4.3. Tolerated use

Reviewers may use a language assistance tool to correct the form of their own report, provided that the text submitted to the tool contains no excerpt from the manuscript and that they inform the editor in the review form.

4.4. Reporting

A reviewer who suspects undisclosed or prohibited use of AI by the authors informs the editor and gives the reasons for this assessment. The reviewer refrains from any definitive judgement based solely on the output of an automated detector.

5. Rules for editors and the journal

5.1. Editorial decisions

Decisions to accept, revise or reject are made exclusively by members of the editorial board. No AI tool is used to draft a decision, a summary of reviews or a letter to authors on substantive matters. Editors are bound by the same confidentiality obligations as reviewers.

5.2. Automated tools used by the journal

The RPM uses similarity detection software during the preliminary screening of manuscripts. It may use automated tools to check references, detect image manipulation or identify possible use of generative AI. Any tool used routinely is disclosed on the journal's website, after having been tested.

5.3. Human oversight

The results of automated tools are indications, not evidence. They are always reviewed by an editor before any decision. No manuscript is rejected solely on the basis of an AI detection score.

5.4. Production and dissemination

The journal may use assistance tools for copyediting, layout, metadata generation or translation of abstracts. These uses are supervised by the editorial team and do not alter the scientific content approved by the authors.

6. Breaches and measures

Any alleged breach is handled in accordance with the COPE flowcharts, with due regard for the right to be heard: the persons concerned are informed and invited to explain themselves before any decision.

Depending on the seriousness and the stage of the process, the journal may take the following measures:

Situation Possible measure
Permitted but undisclosed use, before publication Request for a declaration and, where appropriate, for revision
Prohibited use detected before publication Rejection of the manuscript
Fabricated references or data, before publication Rejection and notification of the authors' institution
Breach discovered after publication Correction, expression of concern or retraction of the article
Reviewer who uploaded a manuscript to an AI tool Report set aside, removal from the list of reviewers
Repeated or serious breach Ban on submissions for a period set by the editorial board

7. Declaration template and policy review

7.1. Declaration template (to be placed before the references)

Declaration of generative AI use

During the preparation of this work, the author(s) used [tool name, version] in order to [specify the use: translation, rewording of passages, help with writing code, etc.], for the following sections: [specify]. After using this tool, the author(s) reviewed and edited the content as needed and take(s) full responsibility for the content of the publication.

Where there is no use covered by point 3.3: "No generative artificial intelligence tool was used in the preparation of this manuscript, beyond spelling and grammar correction."

7.2. French version of the declaration

Déclaration d'utilisation de l'intelligence artificielle

Au cours de la préparation de ce travail, le(s) auteur(s) a (ont) utilisé [nom de l'outil, version] afin de [préciser l'usage], pour les sections [préciser]. Après utilisation de cet outil, le(s) auteur(s) a (ont) vérifié et modifié le contenu autant que nécessaire et assume(nt) l'entière responsabilité du contenu de la publication.

7.3. Review of the policy

Given the rapid evolution of tools and international standards, this policy is reviewed at least once a year by the editorial board. The version in force is the one published on the journal's website on the date the manuscript is submitted.