Generative AI Policy

1) Purpose and scope

Pedagogical Perspective (PedPer) acknowledges that generative AI tools (e.g., large language models, AI-assisted writing/translation tools, and AI image tools) may be used in scholarly workflows. This policy sets requirements for authors, reviewers, editors, and editorial staff to ensure transparency, confidentiality, and research integrity.

2) Core principles

  • Human accountability: Authors, reviewers, and editors remain fully responsible for the accuracy, originality, ethical compliance, and integrity of all content and decisions.
  • Transparency: Permitted AI use must be disclosed as described below.
  • Confidentiality & data protection: Manuscripts and peer-review materials must not be shared with third-party AI systems in ways that breach confidentiality, copyright, or data protection.

A) Policy for Authors

A1) Permitted uses 

Authors may use generative AI tools for:

  • language polishing (grammar, clarity, style),
  • translation support,
  • non-substantive formatting assistance,
  • coding/data-processing assistance (provided outputs are verified and documented),
    provided that authors review all outputs and disclose the use.

A2) Prohibited / unacceptable uses

Generative AI tools must not be used to:

  • fabricate or falsify data, results, participant information, or observations,
  • generate fake or inaccurate references, citations, quotations, or DOIs,
  • produce unattributed paraphrases of copyrighted text (plagiarism),
  • create or modify images/figures in a misleading way or alter the interpretation of results,
  • generate substantial scientific content (e.g., claims, interpretations, conclusions) without appropriate human authorship and without disclosure.

A3) Authorship

AI tools cannot be listed as authors. Only humans can meet authorship criteria and take responsibility for the work.

A4) Mandatory AI Use Disclosure

At submission, authors must include an AI Use Disclosure in the cover letter and/or the manuscript (e.g., Acknowledgements), stating:

  • the tool(s) used (name and version if known),
  • the purpose (e.g., language editing, translation, code assistance),
  • which sections/tasks were supported,
  • confirmation that authors reviewed and verified outputs.

Suggested short disclosure:

“Generative AI tools were used for language editing/translation assistance. The authors reviewed and take full responsibility for the final content.”

Suggested detailed disclosure:

“We used [Tool name] to support [purpose]. AI assistance was limited to [sections/tasks]. All outputs were verified by the authors, who take full responsibility for the integrity and accuracy of the manuscript.”

A5) Privacy and sensitive information

Authors must not upload confidential, proprietary, or personal/sensitive data (including identifiable participant information) to AI tools unless they have a lawful basis and the tool’s terms allow such use. Any AI use involving sensitive data must be clearly described and justified.

A6) Compliance and consequences

PedPer may apply similarity checks and integrity screening. Undisclosed or inappropriate AI use may result in:

  • requests for clarification/correction,
  • rejection,
  • post-publication correction, retraction, or expression of concern,
  • notification to institutions/funders when appropriate.

B) Policy for Reviewers

B1) No AI processing of manuscripts 

Reviewers must not upload, paste, or process any part of the submitted manuscript, supplementary files, or peer-review correspondence in any generative AI tool or third-party system. This protects confidentiality, copyright, and data security.

B2) Reviewer accountability

Peer-review reports must reflect the reviewer’s own expert judgement. AI-generated peer review reports are not acceptable.

C) Policy for Editors and Editorial Office

C1) Confidentiality and decision integrity

Editors and staff must not share manuscripts, reviewer reports, or editorial correspondence containing manuscript content with generative AI tools in ways that breach confidentiality, copyright, or data protection.

C2) Limited permitted uses 

Editors/staff may use AI tools for:

  • language polishing of general editorial communications,
  • workflow support (planning, checklists),
  • metadata consistency checks,
    provided that manuscript content is not shared and editorial decisions remain human-led and documented.

D) Figures, images, and multimedia

  • AI-generated or AI-modified images are permitted only if clearly disclosed and not misleading.
  • Any AI-assisted image creation/modification must be described in the figure legend and/or methods (tool, purpose, and what was changed).
  • Image manipulation that changes the interpretation of results is prohibited.

Related Policies: Publication Ethics & Malpractice • Conflict of Interest Statement • Plagiarism & Similarity Policy • Complaints & Appeals Policy • Handling Allegations of Misconduct Policy • Corrections, Retractions & Expressions of Concern Policy • Thematic Issue Policy