Generative Artificial Intelligence Utilization Policy

MATHEdunesa: Jurnal Ilmiah Pendidikan Matematika recognizes that generative artificial intelligence (AI) tools are increasingly used in research and scholarly writing. The journal supports the responsible, transparent, and accountable use of these tools. At the same time, it protects the integrity, originality, and reliability of the scholarly record.

This policy applies to Authors, Reviewers, Editors, and the editorial team. It follows the position of the Committee on Publication Ethics (COPE) on authorship and AI tools, the recommendations of the International Association of Scientific, Technical and Medical Publishers (STM), and practices widely adopted by internationally reputable, Scopus-indexed journals.

1. Definitions

Generative AI refers to technologies that produce text, images, code, data, or other content in response to prompts. Examples include large language models (LLMs) and chatbots such as ChatGPT, Gemini, Claude, and Copilot, as well as image generators and AI-based writing, paraphrasing, or translation tools that produce new content.

The following are not considered generative AI use under this policy and do not require disclosure:

basic spelling and grammar checkers;
reference management software, such as Mendeley or Zotero;
standard statistical or qualitative analysis software, such as SPSS, R, or NVivo, used in the conventional way; and
mathematical software, such as GeoGebra or computer algebra systems, used for computation or visualization.

2. Core Principles

Human accountability. Authors remain fully responsible for the entire content of their manuscript, including any part produced or assisted by AI.
Transparency. Any substantive use of generative AI must be disclosed.
Human scholarly contribution. AI may assist the research and writing process, but it must not replace the Authors’ own intellectual work, critical thinking, interpretation, or conclusions.
Integrity and confidentiality. AI must not be used in ways that compromise research integrity, data privacy, or the confidentiality of the peer review process.

3. AI and Authorship

Generative AI tools cannot be listed as Authors or co-Authors, and they cannot be cited as authors of any part of the work. Authorship requires human qualities that AI cannot provide: accountability for the work, approval of the final manuscript, the ability to respond to questions about the work, and responsibility for its integrity. This is consistent with the journal’s Authorship and Contributorship Policy.

4. Responsibilities of Authors

4.1 Permitted Uses (with disclosure)

Authors may use generative AI, with appropriate human oversight and disclosure, to:

improve the language, grammar, readability, and clarity of text written by the Authors;
translate text originally written by the Authors, provided the result is carefully checked;
assist with coding or the debugging of analysis scripts, provided the output is verified; and
support brainstorming or the organization of ideas, provided the substantive content, arguments, and interpretations are the Authors’ own.

4.2 Prohibited Uses

Authors must not use generative AI to:

generate substantial portions of the manuscript, such as the literature review, discussion, or conclusions, without meaningful intellectual contribution and critical revision by the Authors;
fabricate, falsify, or manipulate research data, results, participant responses, interview transcripts, or student work;
generate references or citations that the Authors have not personally verified (fabricated or inaccurate references generated by AI are treated as a serious breach of research integrity);
paraphrase other people’s work to disguise plagiarism or avoid similarity detection;
create or alter images, figures, graphs, or photographs of research data or classroom activities, unless AI image generation is itself part of the research design and is fully described in the Methods section; or
upload confidential or identifiable research data, such as student names, scores, interview recordings, or school data, to AI tools that do not guarantee data protection, without appropriate consent and safeguards.

4.3 AI as Part of the Research Design

MATHEdunesa welcomes research that studies generative AI in mathematics education. Examples include studies of students’ or teachers’ use of AI, AI-supported learning designs, and analyses of AI-generated mathematical solutions. In such studies, the use of AI is part of the research itself and must be described in the Methods section. The description should include:

the name and version of the tool and its provider;
the date or period of use;
the prompts or prompt strategies used, provided as supplementary material where appropriate; and
how AI outputs were evaluated, verified, or analyzed.

4.4 Verification

Authors must critically review, edit, and verify all AI-assisted content for accuracy, completeness, and bias. This includes mathematical correctness, the validity of arguments, and the accuracy of every citation. AI tools can produce content that is plausible but incorrect. Responsibility for any such errors rests entirely with the Authors.

4.5 Disclosure Statement

Authors who used generative AI in preparing the manuscript must include a Declaration of Generative AI Use section, placed before the References, using the following format:

During the preparation of this work, the author(s) used [name of tool, version] in order to [purpose, e.g., improve language and readability]. After using this tool, the author(s) reviewed and edited the content as needed and take full responsibility for the content of the publication.

If no generative AI was used, Authors should state:

The authors declare that no generative AI tools were used in the preparation of this manuscript.

The use of AI must also be declared in the Originality Statement submitted through OJS. Disclosure of language-editing use will not, by itself, affect the editorial decision.

5. Responsibilities of Reviewers

Manuscripts under review are confidential documents. Reviewers:

must not upload a manuscript, or any part of it, including the abstract, data, or figures, into generative AI tools, because doing so may breach the Authors’ confidentiality, proprietary rights, and data privacy;
must not use generative AI to write or substantially draft review reports, because peer review requires the reviewer’s own critical judgment and expertise, and the reviewer is fully responsible for the report’s content; and
may use basic language tools to polish the wording of their own review text, provided no manuscript content is entered into the tool.

Reviewers who suspect inappropriate or undisclosed AI use in a manuscript should inform the Editor, giving reasons.

6. Responsibilities of Editors and Editorial Staff

Editors and editorial staff:

must not upload submitted manuscripts, review reports, or editorial correspondence into generative AI tools;
must not use generative AI to make, or substantially draft, editorial evaluations or decisions on manuscripts; and
may use AI-based tools that the journal has formally adopted for technical screening, such as similarity checks, provided these tools protect confidentiality and their results are assessed by a human Editor.

7. Handling Suspected Misuse

AI-detection tools are not fully reliable. MATHEdunesa therefore does not reject manuscripts on the basis of detection scores alone. Concerns about undisclosed or inappropriate AI use are assessed by Editors on the basis of the available evidence. Such evidence may include fabricated references, unverifiable data, inconsistent content, or reviewer reports.

Following COPE guidance, the journal may:

request clarification, drafts, raw data, or other supporting evidence from the Authors;
require revision and proper disclosure;
reject the manuscript; or
after publication, issue a Correction or an Expression of Concern, or retract the article where the misuse undermines the reliability of the work, and notify the Authors’ institution where appropriate.

8. Related Policies and Policy Review

This policy should be read together with the journal’s Authorship and Contributorship Policy, Conflict of Interest and Competing Interests Policy, Publication Ethics, Anti-Plagiarism, Peer Review, and Research Misconduct policies.

Generative AI technology and publishing practice are evolving rapidly. MATHEdunesa will therefore review and update this policy periodically. In cases not covered here, the Editor will exercise professional judgment in accordance with COPE guidance and the protection of the scholarly record. The current approved version remains in force until a revised version is formally adopted.

References:

COPE, Authorship and AI Tools (Position Statement);
STM, Recommendations for a Classification of AI Use in Academic Manuscript Preparation;
ICMJE, Recommendations for the Conduct, Reporting, Editing, and Publication of Scholarly Work, Section on Artificial Intelligence (AI)-Assisted Technology;
WAME, Chatbots, Generative AI, and Scholarly Manuscripts.

Generative Artificial Intelligence Utilization Policy