KI-Policy

Guidelines for the use of generative artificial intelligence for publications with German Medical Science/PUBLISSO

The following guidelines are based on the recommendations of the Committee on Publication Ethics (COPE) [1], the International Committee of Journal Editors (ICJE) [2], the International Association of Scientific, Technical & Medical Publishers (STM) [3], and the European Commission [4].

Use of generative artificial intelligence (AI) in the preparation of manuscript

AI tools must not replace the actual writing process and may only be used as aids, for example, as part of the brainstorming process, research, or to improve the text. The authors are responsible for the accuracy of their manuscript’s content and form. This also applies to the portions generated by AI. It is therefore essential to review all content created using generative AI carefully. Authors are required to ensure that AI-generated statements are adequately supported by appropriate references and that the original sources on which the output is based are cited. AI-generated output itself is not protected by copyright. Authors only obtain copyright protection through significant revision, enhancement, and supplementation of the AI-generated output.

Any use of generative AI in the preparation of manuscripts must be disclosed. This includes both its use during the research process, such as in data analysis, and its use as a tool during the writing process. Depending on the nature and extent of the use of generative AI, the disclosure should be included in the Methods section or as a note at the end of the manuscript. This statement should explain how and for what purpose generative AI was used clearly and in detail.

The following information must be included in the declaration regarding the use of generative AI:

Name of the AI-powered software
Version
Date of use
Scope and nature of use
Prompts, if applicable

Please note the following instructions regarding specific use cases:

Graphical representations of own research processes and results may be created using generative AI. The figure caption must state the name of the AI software used, the version and the date of use. Any use of generative AI to create images or videos exceeding this is prohibited. In all cases, it must be ensured that the terms of use for the generative AI permit publication as part of the article.
If a translation of the manuscript has been produced using generative AI, this must be indicated in the form of a note.
The use of generative AI as a tool for spell-checking and grammar checking, as well as for formatting or citation, is permitted without disclosure.

In addition to careful monitoring and a transparent declaration of how generative AI is used, it is essential to approach the output of generative AI critically and to be aware of the limitations and weaknesses of this type of software. According to the European Commission, these include potential bias in the data used to train generative AI, potential bias in the prompt, inaccuracies or so-called “hallucinations” by generative AI, i.e. the generation of false information, fabricated citations and non-existent references, as well as a lack of transparency regarding the process behind the output (generative AI as a “black box”) [4].

When entering data into generative AI, care must be taken to ensure that no data protection or copyright infringements occur, for example when uploading research data or quotations. Confidential and personal data must not be entered into publicly accessible AI systems. Authors are also obliged to familiarise themselves with the terms of use and privacy policies of any AI system before using it. Furthermore, it is inadvisable to input one’s own manuscript into a generative AI tool that stores and reuses the data for training purposes. Entrusting your own research to AI can lead to a loss of reputation and recognition of your research achievements.

Generative artificial intelligence does not meet the criteria for authorship as defined in the journal policy (https://www.publisso.de/en/publishing/journals/journal-policy) and therefore must not be listed as an author. In particular, as a machine, generative AI cannot bear responsibility for the content and research methodology of the publication. Therefore, the output of generative AI may not be cited as a source, nor are references to sources in which generative AI is listed as an author permitted.

Failure to comply with these guidelines may result in the rejection of the manuscript.

Use of generative AI in the review process

Reviewers and editors must not upload submitted manuscripts or parts of manuscripts to generative AI, as this could compromise the confidentiality of the review process or violate data protection and copyright laws. This rule also applies to cases where the undisclosed use of AI is suspected.

Generative AI must not be used when conducting the review either. The review process is an important quality assurance measure that must be based on human expertise.

For reasons of confidentiality, authors are prohibited from entering the review report submitted by the reviewers into a generative AI system. Revising the manuscript is the responsibility of the authors and must not be done using generative AI.

Use of generative AI by GMS/PUBLISSO

Currently, GMS/PUBLISSO does not use generative AI as a tool, neither in preparing manuscripts for the review process, nor during the editorial processing of accepted manuscripts, nor as part of the technical implementation of the publication. Should this change, GMS/PUBLISSO commits to announcing it.

Note

The English version of this policy was translated with DeepL (https://www.deepl.com/, free version, June 2026) and revised carefully.

References

[1] Committee on Publication Ethics (COPE). COPE Focus on artificial intelligence. Available from: https://publicationethics.org/cope-focus/artificial-intelligence

[2] International Committee of Journal Editors (ICJE). V. Use of Artificial Intelligence in Publishing. In: Recommendations for the Conduct, Reporting, Editing, and Publication of Scholarly Work in Medical Journals. 2026 Jan. Available from: https://www.icmje.org/recommendations/browse/artificial-intelligence/

[3] International Association of Scientific, Technical & Medical Publishers (STM). Generative AI In Scholarly Communications. 2023. Available from: https://stm-assoc.org/document/stm-generative-ai-paper-2023/23/

[4] European Commission. Living guidelines on the responsible use of generative AI in research. Brussels: 2025 Apr 15. Available from: https://research-and-innovation.ec.europa.eu/news/all-research-and-innovation-news/guidelines-responsible-use-generative-ai-research-developed-european-research-area-forum-2024-03-20_en

ZB MED is a member of DataCite

DataCite is an international not-for-profit organisation that offers services and know-how relating to the management, referencing and citation of research data.

ZB MED advocates equality

ZB MED received the TOTAL E-QUALITY award in 2010, 2013, 2017 and 2021 for successfully implementing equality in its staffing structure and personnel policies.

ZB MED advocates Open Access

ZB MED promotes and supports open access to scientific research through the PUBLISSO portal and other services.

ZB MED is sponsered by the Ministry of Culture and Science of North Rhine-Westphalia

The Ministry of Culture and Science of the German State of North Rhine-Westphalia is sponsoring ZB MED.

ZB MED cooperates with the AWMF

The Association of Scientific Medical Societies (AWMF) brings together the interests of medical science and represents them externally. It acts on behalf of more than 180 medical and scientific societies.