Policy on the Use of Artificial Intelligence (AI) in Manuscript Preparation, Peer Review, and Editorial Processes

 

Background

This policy guides the ethical, transparent, and responsible use of artificial intelligence (AI) tools in all aspects of publishing in the Rwanda Journal of Medicine and Health Sciences (RJMHS), it refers to the AJPP’s Guidance on the Use of AI available at https://doi.org/10.4314/gmj.v58i1.1.These policies apply to all authors, reviewers, editors, and journal staff involved in manuscript preparation, submission, review, and publication.

 

1. For Authors

Large Language Models (LLMs) such as ChatGPT or similar tools do not meet authorship criteria and may not be listed as authors.

Authorship requires accountability, which cannot be applied to AI systems.

Any use of AI tools in content generation, data analysis, or drafting must be clearly disclosed in the Methods section (or Acknowledgments if no Methods section exists).

AI-assisted copy-editing (grammar, spelling, punctuation, readability, style) does not require disclosure, provided authors take responsibility for the final text.

In all cases, human authors remain fully accountable for accuracy, originality, and integrity of the work.

 Generative AI Images and Figures

RJMHS does not permit publication of AI-generated images, graphics, or illustrations, due to unresolved copyright, integrity, and ethical concerns.

Exceptions may be considered :

Images from licensed agencies with acceptable legal frameworks.

Images or videos directly referenced in articles specifically about AI, subject to editorial approval.

AI tools developed with verifiable scientific datasets, provided attribution and ethical compliance are ensured.

All permitted exceptions must be clearly labelled as AI-generated in figure legends or captions.

Text-based and numerical display items (tables, flowcharts, simple graphs) are not restricted, but if AI was used to manipulate or enhance such outputs, authors must disclose this in the caption.

Compliance and implications

Failure to disclose AI use or misuse of AI (e.g., undisclosed AI-generated text, images, or peer reviews) may result in manuscript rejection, correction, or retraction.

Cases of misconduct will be escalated to authors’ institutions in line with COPE guidelines.

2. For Reviewers

Peer reviewers are selected for their expertise and remain fully accountable for their evaluations.

Reviewers must not upload manuscripts into generative AI tools, due to confidentiality, data security, and reliability concerns.

If a reviewer uses an AI tool in any way (e.g., for summarisation or cross-checking claims), this must be declared transparently in the review report.

Review reports must remain primarily human-authored, and AI cannot be used to replace critical analysis.

3. For Editors

Editors and staff may use AI for workflow support (e.g., plagiarism checks, metadata extraction, grammar correction).

AI tools must not be used for editorial decision-making or independent peer review.

4. Policy Review

This policy will be reviewed regularly in line with best practices from leading publishers and other agencies (e.g., Springer Nature, COPE, ICMJE,) and adapted as the field evolves.