Policy on the use of Artificial Intelligence

Terra Latinoamericana recognizes the principles established in the Heredia Declaration (https://doi.org/10.15359/ree.28-S.19967) on the responsible use of Artificial Intelligence in scientific communication and promotes their application throughout all stages of the editorial process.

Terra Latinoamericana recognizes that authorship, intellectual responsibility, and accountability for manuscripts belong exclusively to natural persons, in accordance with the Mexican legal framework governing copyright and intellectual property.

1. Definition and scope

This policy applies to all stages of research and manuscript preparation for submissions to Terra Latinoamericana. Generative AI is defined as any tool capable of automatically producing text, images, code, or data.

2. Principles of authorship and responsibility

       i.          Prohibition of AI authorship. Under no circumstances may an AI system be listed as an author or co-author of a manuscript. AI tools do not meet the requirements for authorship, including approval of the final version of the manuscript and legal and ethical responsibility for its content.

     ii.          Human responsibility. Human authors are solely responsible for the content of their manuscripts. Therefore, any errors, biases, plagiarism, inaccuracies, or false information generated using AI -including fabricated references, distorted findings, or fictitious cross-references- shall remain the exclusive responsibility of the authors.

3. Permitted uses and restrictions

       i.          Editing and language improvement. The use of AI tools to improve manuscript readability (including grammar correction, translation, and language editing) is permitted, provided that the original meaning and scientific findings are not altered.

     ii.          Generation of interpretative content, The use of AI tools to support the drafting of interpretative sections must be subject to comprehensive human supervision and validation. Authors shall assume full responsibility for all content generated with such assistance.

   iii.          Restrictions on images and data. The use of AI to generate or manipulate scientific images (including, but not limited to, microscopy images, spectra, radiographs, and similar materials) is strictly prohibited. Likewise, the generation of synthetic datasets is prohibited unless such datasets are clearly identified, methodologically justified, and fully described within the manuscript, or unless the primary objective of the research is the study of AI itself, in which case their use must be rigorously documented.

4. Transparency and disclosure requirements

Any author who has used AI tools at any stage of the research or manuscript preparation process must formally disclose such use. The submission letter must include information provided to the Editor-in-Chief regarding the AI tool used, the purpose of its use, and the sections of the manuscript to which it was applied.

5. Detection and review procedures

       i.          Initial screening. Terra Latinoamericana may employ AI-detection software during the technical pre-screening stage before assigning a manuscript to a Section Editor.

     ii.          Editor-in-Chief assessment. A high AI-detection score will not constitute grounds for automatic rejection. However, it may trigger a manual review process. If the Editor-in-Chief identifies fabricated references, fabricated results, or misleading content generated through AI tools, the manuscript will be rejected on the grounds of scientific misconduct. Such manual evaluations shall be conducted by subject-matter experts to ensure fair judgment beyond automated detection results.

   iii.          Confidentiality during peer review. External reviewers are prohibited from uploading submitted manuscripts to AI tools for analysis, as doing so may violate the confidentiality and intellectual property rights of the authors.

6. Sanctions and detection procedures

       i.          Use of AI detection tools. Results obtained from AI-detection software shall be considered solely as advisory indicators and may lead to additional editorial review.

     ii.          Classification of violations:

a)     Minor ethical violation. Failure to disclose the use of AI for language editing or translation, provided that such use does not affect the originality or validity of the scientific results.

b)     The sanction shall consist of returning the manuscript to the authors for correction and formal disclosure.

c)     Scientific misconduct (Serious Violation). The fabrication, falsification, or alteration of data, results, images, references, or scientific interpretations using AI tools or any other means shall constitute a serious ethical violation.

   iii.          Disciplinary measures. In cases classified as scientific misconduct, Terra Latinoamericana will immediately reject the manuscript and may impose temporary or permanent restrictions on future submissions by the authors involved, depending on the severity of the violation and following evaluation by the Editorial Committee.

   iv.          Protection of confidential information. Authors shall refrain from uploading confidential information, personal data, sensitive information, or materials subject to legal, ethical, or contractual restrictions to publicly accessible AI tools, unless such tools explicitly guarantee adequate privacy and data protection safeguards.

This policy should be periodically reviewed and updated by Terra Latinoamericana in response to technological advances, regulatory developments, and evolving international best practices regarding Artificial Intelligence and research integrity.