Clinical & Molecular Biomedicine (CMB) recognizes the growing role of artificial intelligence (AI), generative AI, large language models, and AI-assisted digital tools in research, writing, editing, and scholarly communication. The journal supports the responsible use of AI tools when they are used transparently, ethically, and under full human oversight.

AI tools may assist authors, reviewers, and editors, but they must not replace human expertise, scientific judgment, accountability, or ethical responsibility.

The use of AI in manuscripts submitted to CMB must follow the principles of:

• Transparency

• Human accountability

• Accuracy

• Confidentiality

• Research integrity

• Ethical compliance

• Proper attribution

Authors remain fully responsible for all content submitted to the journal, including any text, data, figure, analysis, citation, or interpretation generated or assisted by AI tools.

AI tools, including large language models and generative AI systems, cannot be listed as authors or co-authors.

Authorship requires legal, ethical, and scholarly accountability. AI tools cannot approve the final version of a manuscript, respond to editorial questions, take responsibility for research integrity, or be held accountable for errors or misconduct.

Only human contributors who meet the journal’s authorship criteria may be listed as authors.

AI tools may be used responsibly for limited support tasks, provided that authors carefully review and verify all outputs.

Permitted uses may include:

• Improving grammar, spelling, and readability

• Language polishing

• Formatting assistance

• Organizing text structure

• Summarizing author-generated notes or drafts

• Assisting with literature organization

• Supporting data analysis when clearly described in the Methods section

• Use as part of a validated research methodology, where appropriate

AI-assisted outputs must never be submitted without careful human review.

Basic spelling, grammar, punctuation, formatting, and readability checks do not normally require disclosure, provided that the tool is used only to improve language clarity and does not generate new scientific content, analysis, interpretation, or conclusions.

Authors remain responsible for ensuring that such edits accurately reflect their original work.

Authors must disclose the use of AI tools when such tools are used for manuscript preparation, drafting assistance, text generation, summarization, data analysis, image processing, reference organization, or any other task that may affect the intellectual or scientific content of the manuscript.

Disclosure should be included in the Methods, Acknowledgements, or another appropriate section of the manuscript.

The disclosure should include:

• Name of the AI tool or service used

• Version, where available

• Purpose of use

• Section or stage of the manuscript where the tool was used

• Confirmation that the authors reviewed, edited, and approved the final content

Authors may use the following format when disclosing AI use:

“During the preparation of this manuscript, the author(s) used [name of AI tool/service] for [purpose of use]. After using this tool/service, the author(s) reviewed, edited, and verified the content as needed and take full responsibility for the accuracy, integrity, and originality of the final manuscript.”

If AI tools were used as part of the research methodology, authors must describe the use in sufficient detail to allow evaluation and reproducibility.

Authors must critically verify all AI-assisted content before submission.

This includes checking:

• Scientific claims

• Data interpretation

• Statistical outputs

• Citations and references

• Methods descriptions

• Figure legends

• Tables

• Conclusions

AI tools may produce inaccurate, biased, incomplete, misleading, or fabricated information. Authors are responsible for correcting such issues before submission.

AI tools must not be relied upon as authoritative sources for references.

Authors must verify all references against original sources. Fake, inaccurate, incomplete, or misleading references generated by AI tools are considered a serious concern and may lead to editorial queries, manuscript rejection, correction, or further investigation.

AI tools must not be cited as authors of scholarly sources.

The following uses of AI are strictly prohibited:

Data Fabrication

AI must not be used to generate, invent, simulate, or manipulate research data in a way that misrepresents real findings.

Clinical or Experimental Result Fabrication

AI must not be used to create false clinical results, laboratory results, patient outcomes, experimental observations, or representative datasets.

Reference Fabrication

AI must not be used to generate fake, misleading, unverifiable, or non-existent citations.

Image or Figure Misrepresentation

AI must not be used to create, alter, enhance, remove, obscure, or introduce specific features in research images or figures in a way that misrepresents the original data.

Plagiarism or Text Misappropriation

AI must not be used to reproduce copyrighted, confidential, or third-party material without proper permission and attribution.

Peer Review Manipulation

AI must not be used to create fraudulent reviewer responses, automated peer feedback, fake reviewer identities, or any material intended to manipulate the peer review process.

The use of generative AI or AI-assisted tools to create or modify figures, images, graphical abstracts, clinical images, microscopy images, or other visual material is generally not permitted.

An exception may be made when AI tools are used as part of the research methodology, such as biomedical image analysis, computational imaging, or validated AI-assisted diagnostic methods.

In such cases, authors must provide full transparency by clearly describing:

• The AI tool, model, or software used

• Version details

• Manufacturer or developer, where applicable

• Technical parameters

• How the tool was used

• How the output was verified

• Whether original or pre-processed image files are available for editorial review

This information must be included in the Methods section.

AI-generated artwork, including graphical abstracts or illustrations, requires prior editorial approval and must be clearly disclosed.

Authors must not upload confidential, unpublished, copyrighted, sensitive, identifiable, or patient-related information into public or external AI tools unless they have confirmed that the tool protects privacy, confidentiality, intellectual property, and data security.

This includes:

• Unpublished manuscripts

• Reviewer comments

• Patient information

• Clinical images

• Personal data

• Institutional confidential data

• Copyrighted third-party material

• Proprietary datasets

Authors are responsible for reviewing the terms and conditions of any AI tool before use.

Reviewers must maintain the confidentiality of manuscripts and peer review materials.

Reviewers must not upload submitted manuscripts, figures, data, supplementary files, or confidential review materials into public or external AI tools.

If a reviewer uses an approved AI-assisted tool for limited support, such as improving the clarity of their own review text, they must ensure that no confidential manuscript content is disclosed and must declare any substantive AI assistance to the editor.

Reviewer evaluations must reflect the reviewer’s own expert judgment.

Editorial decisions must be made by human editors.

AI tools may support administrative or editorial workflow tasks only when used responsibly, securely, and under human oversight. AI must not replace editorial judgment, peer review assessment, or final decision-making.

Editors must protect the confidentiality of submitted manuscripts and must not use public AI tools in a way that exposes confidential manuscript content, author information, reviewer identity, or editorial correspondence.

Failure to disclose relevant AI use may be treated as a breach of publication ethics.

Depending on the nature and severity of the case, the journal may request clarification, require correction of the manuscript, reject the submission, publish a correction, issue an expression of concern, or initiate further investigation according to the journal’s publication ethics procedures.

Because AI technologies and publishing standards are developing Updates**

Because AI technologies and publishing standards are developing rapidly, CMB may update this policy from time to time. Authors, reviewers, and editors are encouraged to review the current policy before submission, review, or editorial handling.