Use of Generative AI

Responsible Use of Artificial Intelligence
Use of Generative AI
Transparency, Human Responsibility, Confidentiality, and Research Integrity

General Policy

Notulae Scientia Biologicae recognizes that artificial intelligence and generative AI tools may support certain aspects of research communication, language improvement, data processing, and manuscript preparation.

Their use must be transparent, responsible, lawful, and consistent with research integrity, confidentiality, authorship, copyright, data protection, and ethical requirements.

Human authors remain fully responsible for all submitted and published content, regardless of whether AI tools were used.

AI Tools and Authorship

Artificial intelligence tools, language models, chatbots, software systems, and other non-human technologies cannot be listed as authors or co-authors.

Authorship requires accountability, approval of the final work, responsibility for the integrity of the research, and the capacity to respond to questions, corrections, or ethical concerns. These responsibilities can be assumed only by human authors.

Authors are responsible for:

  • the accuracy and originality of the manuscript;
  • the validity of data, analyses, interpretations, and conclusions;
  • the accuracy and authenticity of references and citations;
  • compliance with copyright, privacy, confidentiality, and ethical requirements;
  • disclosure of relevant AI use where required.

Permitted and Responsible Uses

AI-assisted tools may be used responsibly for limited support tasks, provided that authors carefully review, verify, and take full responsibility for the resulting content.

Language Improvement
Grammar, spelling, readability, and stylistic refinement.
Translation Support
Assistance with translation, subject to careful human verification.
Technical Assistance
Coding, formatting, structured data preparation, or analytical support, where appropriate and transparently documented.
Idea Organization
Support for outlining, structuring, or organizing text, without replacing the authors’ scientific judgment.

Routine use of spelling, grammar, reference-management, formatting, or standard editing software does not normally require disclosure.

Disclosure of AI Use

Substantive use of generative AI in preparing, analyzing, interpreting, or presenting the work must be disclosed clearly and accurately.

Disclosure is generally expected when AI tools have materially contributed to:

  • drafting or substantially rewriting manuscript content;
  • generating or modifying figures, images, tables, or visual material;
  • creating, processing, or interpreting datasets;
  • developing code, models, algorithms, or analytical workflows;
  • generating summaries, interpretations, or scientific conclusions.

The disclosure should identify the tool, provider, version where available, purpose of use, and the nature of the authors’ verification.

A suitable statement may be included in the Methods, Acknowledgements, or another appropriate section:

During the preparation of this manuscript, the authors used [tool and version] for [specific purpose]. The authors reviewed and verified all AI-assisted content and take full responsibility for the final manuscript.

Prohibited Uses

AI tools must not be used in ways that compromise scientific integrity, transparency, accuracy, confidentiality, or accountability.

  • fabrication or falsification of data, results, analyses, images, references, approvals, permits, or scientific claims;
  • creation of false or unverifiable citations;
  • undisclosed generation or manipulation of images or visual evidence;
  • replacement of human scientific judgment or responsibility;
  • creation of false reviewer identities or review reports;
  • manipulation of peer review or editorial communication;
  • uploading confidential, personal, restricted, proprietary, or unpublished information to unauthorized external AI platforms.

References and Citations

Authors must verify every citation and bibliographic reference against the original source.

AI-generated references, fabricated citations, incorrect DOIs, false publication details, or references that the authors have not verified are unacceptable.

AI systems should not be cited as primary scholarly sources unless the use of a specific AI system is itself the subject of the research and the citation is methodologically justified.

AI-Generated Data, Images, and Scientific Content

AI-generated or AI-modified data, images, figures, tables, code, interpretations, or scientific content must not be presented as authentic research observations unless their origin, generation, validation, and limitations are transparently described and scientifically justified.

Authors must retain original source data, unprocessed images, prompts, code, settings, or other relevant records where AI-assisted processing materially affects the work.

The journal may request such material during editorial assessment or after publication.

Confidentiality and Data Protection

Authors must not upload confidential, identifiable, restricted, proprietary, unpublished, or ethically sensitive information to AI systems unless they are authorized to do so and the platform provides appropriate safeguards.

This includes:

Personal or medical information
Unpublished research data
Confidential manuscripts
Proprietary or restricted information
Reviewer reports and editorial correspondence
Copyright-protected third-party material

Use of AI by Reviewers

Manuscripts, supplementary files, review invitations, reviewer reports, and editorial correspondence are confidential.

Reviewers must not upload manuscript content to public or unauthorized AI platforms because doing so may breach confidentiality, copyright, privacy, and data-protection obligations.

Reviewers remain personally responsible for the accuracy, independence, confidentiality, and scholarly quality of their reports.

AI tools must not replace the reviewer’s expert evaluation or be used to generate a review that the reviewer has not personally verified and accepted.

Use of AI by Editors

Editors must protect the confidentiality of submitted manuscripts, reviewer identities, editorial correspondence, and unpublished information.

Confidential material must not be uploaded to unauthorized AI platforms.

AI-assisted tools may support limited administrative or technical tasks, but editorial decisions must remain based on human scholarly judgment, peer-review evidence, ethical requirements, and the journal’s editorial criteria.

AI-detection tools and automated indicators must not be treated as conclusive evidence of misconduct. Any concern must be assessed by qualified editors using the available evidence and due process.

Non-Compliance

Undisclosed, misleading, inappropriate, or unethical use of AI may result in editorial action.

Depending on the seriousness and stage of the case, the journal may:

  • request clarification or disclosure;
  • request correction or supporting documentation;
  • suspend manuscript evaluation;
  • reject the manuscript;
  • withdraw an acceptance decision;
  • publish a correction or expression of concern;
  • retract a published article;
  • notify relevant institutions or authorities where appropriate.

Related Policies