Artificial Intelligence (AI) Policy

Guidelines for Authors, Reviewers, and Editors

Forum Geografi recognises that generative artificial intelligence (AI) and AI-assisted technologies can support legitimate scholarly work, while also posing risks to research integrity, confidentiality, and accountability. This policy explains how such technologies may and may not be used at every stage of the publication process. It applies to all manuscripts submitted to the journal and complements the journal’s Publication Ethics—all parties involved, including the authors, peer reviewers, and editors, should comply with this policy.

Forum Geografi focus on spatial and regional analyses that apply GIS, remote sensing, machine learning, and related computational methods, therefore, this policy distinguishes two different uses of AI: (a) generative AI used to help prepare a manuscript (for example, language editing or drafting), which is governed by the disclosure rules below; and (b) AI or machine-learning techniques used as part of the research methodology (for example, image classification or predictive modelling), which are legitimate research methods and must be reported reproducibly in the Methods section rather than treated as generative-AI assistance.

1. General Principles

  • AI cannot be an author. AI tools do not meet authorship criteria because they cannot take responsibility for the work, declare conflicts of interest, or hold copyright. They must not be listed as authors or co-authors.
  • Human accountability. Authors, reviewers, and editors remain fully responsible for the accuracy, integrity, and originality of all content they submit, evaluate, or approve — including any portion produced with AI assistance.
  • Any use of generative AI in preparing a manuscript or a review must be disclosed clearly and honestly.
  • Unpublished manuscripts, figures, data, and review reports are confidential and must not be uploaded to public or third-party AI tools.
  • Where AI or machine learning is used as a research method, it must be described in sufficient detail to allow the work to be reproduced and independently assessed.
  • Data openness. All data sources and the analytical source code should be provided opened/attached as supplementary material for further inspection and verification to confirm there is no data falsification and fabrication. Authors are encouraged to use a public repository with DOI such as Zenodo (https://zenodo.org/) or Open Science Framework (https://osf.io/).

2.  Policy for Authors

2.1  Permitted uses (with human oversight)

Authors may use generative AI tools for the following purposes, provided the output is checked and the author takes full responsibility for the final content:

  1. Language and readability. Improving the overall fluency and readability of the manuscript, and identifying and correcting grammatical and spelling errors.
  2. Reference management. Assisting with citation management, including generating bibliographies and formatting references according to the required style.
  3. Presentation and data handling. Supporting the clear, accurate, and concise presentation of findings, and assisting with the processing and analysis of complex datasets — subject to verification and, where used as a research method, full reporting in the Methods section (see Section 2.4).

2.2  Prohibited uses

The following uses of generative AI are not permitted:

  1. Generating core scholarly content. Producing the primary sections of the manuscript — such as the introduction, methodology, results, or conclusion — which must reflect the authors’ own intellectual contribution and analysis.
  2. Compromising data. Fabricating, altering, or falsifying data or research results.
  3. Claiming authorship. Listing AI tools as authors or co-authors, in the text or in the references.
  4. Manipulating images. Creating, altering, or manipulating visual elements or images in submitted manuscripts, except where AI/ML is itself the research method being reported (see Section 2.4 and Section 5).

2.3  Disclosure requirement

Authors must disclose any use of generative AI in the preparation of the manuscript and in Generative AI Use Declaration Form provided by the journal. The disclosure should name the tool used, the purpose of its use, and confirm author oversight. Place it in a brief statement at the end of the manuscript, before the reference list (a Methods or Acknowledgements statement is also acceptable). Routine spelling/grammar checkers and reference managers without generative features do not require disclosure. A suggested wording is:

Declaration of Generative AI and AI-assisted technologies

During the preparation of this work the author(s) used [NAME OF TOOL / SERVICE] in order to [REASON]. After using this tool/service, the author(s) reviewed and edited the content as needed and take(s) full responsibility for the content of the publication.

2.4  AI and machine learning as a research method

When AI or machine-learning techniques form part of the study itself — for example, land-use/land-cover classification, change detection, hotspot or pattern analysis, downscaling, or predictive modelling — their use is a legitimate and welcome research method, not “generative-AI assistance.” Such methods must be reported transparently and reproducibly in the Methods section, including: the model or algorithm and its version or implementation; key parameters and settings; the training, validation, and input data (with sources); and the accuracy or validation procedure used. Maps, classified rasters, and other analytical figures produced by these methods are research outputs and are permitted, provided they are clearly described in the text and figure captions. This is distinct from the generative creation or cosmetic manipulation of images, which remains prohibited (see Section 5).

3.  Policy for Reviewers

Confidentiality first. A manuscript under review is a confidential document. Reviewers must not upload any part of a manuscript, its figures, tables, data, or supplementary materials to AI tools, as doing so may breach the confidentiality and proprietary rights of the authors and, where personal data are involved, data-protection obligations.

AI does not replace expert judgement. Peer review depends on the reviewer’s scholarly expertise and critical thinking, which AI cannot provide; AI output may also be inaccurate, incomplete, or biased. Reviewers may use AI tools only to lightly improve the language of their own review report, and only if no confidential content from the manuscript is entered into the tool. Any such assistance should be declared to the editor. Reviewers remain fully responsible for the content, accuracy, and fairness of their reports.

4.  Policy for Editors

Confidentiality and human decisions. Editors must keep submitted manuscripts and all related correspondence confidential and must not upload manuscripts, reviewer reports, or decision letters to public AI tools. Editorial evaluation and the final decision on a manuscript must be made by the human editor and must not be delegated to generative AI.

Legitimate editorial tools and oversight. Editors may use journal- or publisher-integrated tools (for example, similarity and plagiarism screening) that operate under appropriate confidentiality safeguards. Where an editor suspects undisclosed or non-compliant AI use, the matter should be handled proportionately under the journal’s Publication Ethics and COPE guidance; a suspicion of AI use is not, by itself, grounds for rejection — the focus is on genuine breaches such as fabrication, plagiarism, or fabricated citations.

5.  Images, Figures, and Visual Materials

The generative creation or alteration of images using AI — including adding, removing, relocating, or obscuring features — is not permitted. Standard, non-deceptive adjustments (such as brightness, contrast, or colour balance applied uniformly) remain acceptable provided no information is obscured or introduced. As stated in Section 2.4, maps and analytical figures generated by AI/ML methods that are part of the study’s methodology are permitted and must be described in the Methods and in the figure captions.

6.  Data Integrity and Accountability

AI must never be used to fabricate, falsify, or selectively alter data or results. References, facts, and quotations produced or suggested by AI must be independently verified, as AI can generate plausible but non-existent citations. In all cases, the human authors, reviewers, and editors bear full responsibility for the integrity of the work. Breaches of this policy will be addressed in accordance with the journal’s Publication Ethics and the relevant COPE guidance, and may result in rejection, correction, or retraction.

7.  Alignment, Effective Date, and Review

This policy aligns with the guidance of the Committee on Publication Ethics (COPE), and complements the Forum Geografi Publication Ethics. Because AI technologies and community norms evolve rapidly, this policy will be reviewed and updated periodically. Authors and reviewers should also consult the journal’s current Author Guidelines.

Version 1.0 · 2026. Effective from the date of publication on the journal website. © Forum Geografi. Distributed under CC BY 4.0.