Artificial Intelligence (AI) Policy

1. Purpose

The Journal of Research in Agriculture and Food Sciences (JRAFS) recognizes that artificial intelligence (AI), including generative artificial intelligence (GenAI), large language models (LLMs), chatbots, AI-assisted writing tools, image-generation tools, coding assistants, and other AI technologies can provide useful assistance in scholarly research, manuscript preparation, editing, analysis, and publishing.

At the same time, the use of AI in scholarly publishing raises important concerns regarding accuracy, originality, plagiarism, fabrication, falsification, copyright, confidentiality, transparency, authorship, accountability, and research integrity.

This policy establishes the principles governing the responsible, ethical, and transparent use of AI in manuscripts submitted to and published by JRAFS.

The policy applies to authors, reviewers, editors, editorial board members, and editorial and publishing staff.

2. General Principles

AI-assisted technologies may be used when their use is scientifically and ethically appropriate, transparent, and consistent with responsible research and publication practices.

The following principles apply:

• AI systems cannot be listed as authors or co-authors.

• Human authors remain fully responsible for the accuracy, originality, integrity, and ethical compliance of their work.

• Authors must carefully review and verify all AI-generated or AI-assisted content before submission.

• AI must not be used to fabricate or falsify data, results, references, images, observations, or other scientific evidence.

• AI-assisted content must not introduce plagiarism, copyright infringement, or misleading information.

• The use of AI must not compromise the confidentiality of unpublished manuscripts, research data, personal information, or peer-review materials.

• Material use of AI must be disclosed in accordance with this policy.

• AI tools must not replace human scientific judgment or editorial decision-making.

3. Acceptable Use of AI by Authors

Authors may use AI-assisted technologies for appropriate purposes, including:

• grammar, spelling, punctuation, and language improvement;

• improving clarity, readability, and organization;

• translation and language assistance;

• suggesting alternative wording or manuscript structure;

• developing literature-search strategies;

• generating preliminary ideas or outlines;

• assisting with coding or programming, provided that the resulting code and outputs are independently verified;

• assisting with data analysis or visualization, provided that the methods, outputs, and conclusions are independently verified;

• assisting with the preparation of figures, graphical abstracts, or other visual materials where permitted by applicable requirements.

The use of AI for these purposes does not transfer responsibility from the authors to the AI system.

Authors remain responsible for the final content of the manuscript and for ensuring that all scientific claims, analyses, references, and conclusions are accurate.

4. AI-Generated or AI-Assisted Scientific Content

Authors must exercise particular caution when AI is used to generate or substantially modify scientific content.

AI systems can generate inaccurate, incomplete, biased, or fabricated information, including non-existent references and incorrect scientific statements.

Authors must independently verify, as applicable:

• scientific facts and statements;

• references and citations;

• numerical results;

• statistical analyses;

• tables and figures;

• data interpretation;

• conclusions;

• quotations and attributions;

• information concerning humans, animals, patients, research participants, or other subjects.

AI must not be used to fabricate, falsify, manipulate, or invent research data, experimental results, clinical findings, statistical results, references, or other scientific evidence.

5. AI and Authorship

AI systems, including ChatGPT and other generative AI tools, must not be listed as authors or co-authors.

Authorship requires accountability for the published work. AI systems cannot assume responsibility for the accuracy, integrity, originality, conflicts of interest, copyright, or ethical aspects of a manuscript.

All listed authors must meet the journal's authorship requirements and remain responsible for the entire manuscript, including any content produced or modified with AI assistance.

6. Disclosure of AI Use

Authors must disclose material use of AI-assisted technologies when AI has contributed substantively to the preparation, analysis, generation, or presentation of a manuscript.

The disclosure should identify:

1. the AI tool or system used;

2. the purpose for which it was used; and

3. where appropriate, the nature or extent of its contribution.

A suitable disclosure may be stated as follows:

Artificial Intelligence Disclosure: During the preparation of this manuscript, the authors used [name of AI tool] for [describe purpose]. The authors reviewed and verified all AI-assisted content and take full responsibility for the accuracy, integrity, and originality of the submitted work.

When AI is used for data analysis, data processing, image generation or modification, statistical analysis, coding, or another substantive scientific procedure, its use should also be described in the relevant Materials and Methods or equivalent section.

Routine use of conventional spelling, grammar, or formatting software that does not generate substantive content generally does not require a separate AI disclosure. Authors nevertheless remain responsible for the final manuscript.

7. AI-Generated Images, Figures, and Visual Materials

Authors must disclose the use of generative AI to create, substantially modify, or enhance images, figures, graphical abstracts, or other visual materials when such use could affect the representation or interpretation of the scientific work.

AI must not be used to create or modify scientific images in a manner that misrepresents research findings.

In particular, AI must not be used to:

• fabricate experimental images;

• add, remove, or alter research findings;

• manipulate microscopy, radiography, histology, electrophoresis, or other scientific images misleadingly;

• create apparently authentic experimental results that did not occur;

• selectively alter visual evidence in a manner that changes its scientific interpretation.

Authors must ensure that AI-generated or AI-modified visual materials comply with applicable copyright, licensing, ethical, and research-integrity requirements.

Editors may request information concerning the origin, source files, processing history, or methods used to create or modify an image or figure.

8. AI in Data Analysis and Research

AI may be used as a research or analytical tool when scientifically justified and appropriately validated.

Where AI contributes materially to research or data analysis, authors should provide sufficient information for editors and readers to understand its role, including, where relevant:

• the AI system, model, or software used;

• version or model information;

• the purpose of the AI application;

• relevant parameters, prompts, or settings where scientifically necessary;

• the type of data processed;

• validation and quality-control procedures; and

• how the outputs were independently verified.

AI-generated outputs must not be assumed to be accurate or authoritative without appropriate verification.

9. Fabrication, Falsification, and Misrepresentation

Using AI to fabricate, falsify, manipulate, or misrepresent research constitutes a serious breach of publication ethics.

Examples include:

• fabricated data;

• fabricated references;

• fabricated quotations;

• fabricated experimental observations;

• invented statistical results;

• manipulated scientific images;

• false descriptions of methods or materials;

• presenting AI-generated information as verified scientific evidence when it has not been verified.

Such conduct may result in rejection, correction, retraction, or other appropriate editorial action in accordance with the journal's publication ethics policies.

10. Plagiarism, Copyright, and Intellectual Property

Authors remain responsible for ensuring that AI-assisted content does not infringe copyright or other intellectual property rights.

AI systems may generate material that resembles existing copyrighted or otherwise protected content. Authors must therefore:

• verify the originality of submitted content;

• provide appropriate citations and attribution;

• obtain permission where required;

• avoid reproducing copyrighted material without authorization;

• verify the authenticity and accuracy of AI-generated references; and

• ensure that AI-generated content does not replace proper citation of the original scientific literature.

The use of an AI system does not exempt authors from normal requirements concerning attribution, citation, permissions, copyright, or intellectual property.

11. Confidentiality and Sensitive Information

Authors, reviewers, editors, and editorial staff must exercise particular caution when using AI systems.

Unpublished manuscripts, confidential peer-review materials, proprietary information, personal data, patient or participant information, unpublished research data, or other confidential material must not be entered into public or third-party AI systems unless adequate confidentiality and data-protection safeguards are in place and such use is permitted.

Reviewers and editors must not upload submitted manuscripts or confidential peer-review information to AI systems where confidentiality cannot be adequately assured.

The confidentiality and integrity of the peer-review process must be maintained at all times.

12. AI Use by Peer Reviewers

Peer reviewers may use AI-assisted tools only in ways that are consistent with the journal's confidentiality, ethical, and security requirements.

Reviewers must not upload a submitted manuscript, peer-review report, unpublished data, or other confidential material into an AI system where confidentiality cannot be assured.

AI must not replace the reviewer's independent scientific judgment.

Reviewers remain responsible for evaluating the manuscript's:

• scientific validity;

• originality;

• methodology;

• interpretation;

• ethical compliance; and

• overall quality.

Any material use of AI in preparing a peer-review report should be disclosed to the journal when requested.

13. AI Use by Editors and Editorial Staff

Editors and editorial staff may use appropriate AI-assisted tools for administrative or editorial purposes, provided that such use does not compromise:

• confidentiality;

• editorial independence;

• research integrity;

• data protection;

• authors' rights; or

• the integrity of the peer-review process.

AI may assist with tasks such as language improvement, formatting, administrative processing, or identification of potential issues.

However, substantive editorial decisions must remain under human responsibility. AI tools must not independently determine acceptance, rejection, retraction, or other substantive editorial outcomes.

14. Editorial Screening and AI Detection

JRAFS may use appropriate software and AI-assisted technologies as part of manuscript screening, quality control, plagiarism assessment, image screening, and research-integrity assessment.

As part of its editorial screening process, JRAFS may use the AI Content Detector powered by Winston to identify manuscripts that may contain substantial AI-generated or AI-assisted content.

AI-detection results are considered screening indicators only. They will not, by themselves, be treated as conclusive evidence of inappropriate AI use, plagiarism, academic misconduct, or research misconduct.

AI-detection systems may produce both false-positive and false-negative results. Accordingly:

• a high AI-detection score does not necessarily establish inappropriate AI use;

• a low AI-detection score does not establish that AI was not used; and

• editorial decisions will not be based solely on an automated AI-detection score.

Where screening raises reasonable concerns, the editorial team may conduct additional assessment, request clarification or disclosure from the authors, examine supporting materials, or take other appropriate editorial action.

The final determination remains the responsibility of the journal's editors.

15. Failure to Disclose Material AI Use

Failure to disclose material AI use when disclosure is required may constitute a breach of the journal's publication policies.

Where appropriate, the editorial team may request an explanation or additional information from the authors.

Depending on the circumstances and seriousness of the issue, possible actions include:

• requesting clarification or disclosure;

• requiring revision of the manuscript;

• rejecting the manuscript;

• publishing a correction or clarification;

• issuing an Expression of Concern; or

• retracting a published article in serious cases.

The action taken will depend on the nature, extent, and consequences of the undisclosed use.

16. Responsibility of Authors

By submitting a manuscript to JRAFS, authors confirm that they:

• have reviewed the entire manuscript;

• accept responsibility for the submitted work;

• have appropriately reviewed and verified AI-assisted content;

• have disclosed material AI use where required;

• have verified references and citations;

• have not used AI to fabricate or falsify scientific information;

• have complied with applicable copyright and intellectual property requirements;

• have not improperly disclosed confidential or sensitive information to AI systems; and

• remain fully responsible for the accuracy, integrity, originality, and ethical compliance of the manuscript.

17. Editorial Discretion

JRAFS reserves the right to request additional information concerning the use of AI during submission, peer review, production, or after publication.

Authors may be asked to provide details regarding AI-generated text, images, data analysis, coding, figures, or other AI-assisted material where necessary to evaluate transparency, research integrity, or compliance with this policy.

The journal may take appropriate action if information provided by authors is found to be incomplete, misleading, or inconsistent with the submitted work.

18. Post-Publication AI Concerns

If concerns regarding AI use arise after publication, JRAFS may investigate the matter in accordance with its publication ethics policies.

Depending on the findings, the journal may issue:

• a correction;

• a clarification;

• an Expression of Concern;

• a retraction; or

• another appropriate post-publication notice.

The objective is to ensure that the published scholarly record remains accurate, transparent, and reliable.

19. Policy Updates

AI technologies and standards for responsible AI use in scholarly publishing continue to evolve. JRAFS may periodically update this policy to reflect developments in:

• research integrity;

• publication ethics;

• copyright and intellectual property;

• data protection and confidentiality;

• scholarly communication; and

• recognized guidance on responsible AI use.

Authors, reviewers, editors, and editorial board members are expected to comply with the version of the policy applicable to their involvement in the publication process.

20. Applicable Ethical Guidance

This policy is informed by internationally recognized principles and guidance concerning responsible AI use in scholarly publishing, including relevant recommendations of the International Committee of Medical Journal Editors (ICMJE) and guidance from the Committee on Publication Ethics (COPE).

JRAFS will continue to review developments in AI and scholarly publishing and will update its practices where necessary to maintain high standards of transparency, accountability, confidentiality, and research integrity.