This policy governs the use of artificial intelligence (AI), including generative language models and automated tools, in the writing, editing, and peer review of manuscripts submitted to Eureka Herba Indonesia. It applies to authors, members of the editorial team, and reviewers alike, and takes effect for all manuscripts under consideration from the date of publication of this policy.

An AI tool cannot be listed as an author or co-author of any article published in this journal. Authorship entails responsibilities that no AI system can discharge: approving the final version of the work, disclosing competing interests, and accepting accountability for the accuracy and integrity of the work. Any manuscript naming an AI tool among its authors will be returned without review.

Authors must disclose any material use of AI in the preparation of a manuscript. The disclosure must identify the tool and its version, the sections of the manuscript affected, and the purpose for which it was used. Disclosure is made in the Artificial Intelligence Use Statement, which is a required element of every article.

Disclosure is required where AI has been used for, among other things: drafting or rewriting any part of the text; generating or editing figures, images, or illustrations; generating, cleaning, or transforming data; writing or generating analysis code; conducting literature searches that shaped the argument; translating substantial passages; or summarising sources.

Routine tools that do not alter scholarly content need not be disclosed: spelling and grammar checkers, reference managers, plagiarism detection software, and basic statistical software operated by the authors.

AI must not be used to fabricate or falsify data, results, images, references, citations, peer reviews, identities, informed consent, or ethical approval. Any such use constitutes a breach of academic integrity and is handled under the journal's policy on allegations of research misconduct.

The use of AI does not diminish the authors' responsibility for the originality, accuracy, confidentiality, and integrity of their work. Authors remain fully accountable for every statement, datum, figure, and reference in the article, including those produced with the assistance of AI, and are expected to verify each of them independently.

Members of the editorial team may use AI only for administrative and technical tasks, such as checking the completeness of a submission against the author guidelines, language checking, or similarity screening.

AI must not be used to replace scholarly judgement or editorial decision-making. Decisions to accept, revise, or reject a manuscript rest on the academic judgement of the editors and on the reports of reviewers, and on nothing else.

Editors must not upload unpublished manuscripts, author identities, or review reports into public AI applications. Such material is confidential, and uploading it would place it beyond the journal's control. The use of AI by an editor does not diminish that editor's responsibility for the decisions taken.

Reviewers for Eureka Herba Indonesia are prohibited from using artificial intelligence in the peer review process. The prohibition is absolute and is not conditional on the tool used or on the safeguards claimed for it. Specifically, a reviewer must not:

• use AI to analyse or evaluate a manuscript;
• use AI to draft, expand, or refine a review report, in whole or in part; or
• enter or upload a manuscript, or any part of a manuscript, into any AI application.

A manuscript under review is a confidential document belonging to its authors. Entering it into an AI application places it outside the journal's control and may expose unpublished work to third parties, irrespective of any assurance given by the operator of the tool.

On submitting a review, reviewers attest that they have complied with this prohibition. A breach results in the review being set aside, the manuscript being reassigned, and the reviewer being removed from the journal's panel.

Every article published in this journal carries an Artificial Intelligence Use Statement, placed with the other declarations at the end of the article. The statement is mandatory whether or not AI has been used; where it has not been used, that fact is stated explicitly. Model wording is given in the journal's declaration templates.

Suspected breaches of this policy are handled under the journal's procedure for allegations of research misconduct: preliminary examination of the evidence, a request for clarification, investigation, a documented decision, and corrective action. Depending on the gravity of the breach and the stage at which it is detected, corrective action may take the form of a request for revised disclosure, rejection of the manuscript, a published correction, or retraction of a published article.

Given the pace at which these tools are developing, this policy is reviewed at least annually by the editorial board and revised as required. The version in force at the time a manuscript is submitted applies to that manuscript.