What Should Researchers Do When a Journal Has No Clear AI Policy?

What Does It Mean When the Journal Says Nothing About AI?

You are ready to submit a manuscript. You check the author guidelines for generative AI, ChatGPT, large language models, and AI-assisted writing, but find no clear policy.

Does that mean AI use is allowed? Prohibited? Must it be disclosed anyway?

None of those conclusions necessarily follows from silence. The journal may follow a publisher-wide policy located elsewhere, rely on broader authorship or research-integrity requirements, or simply have no explicit rule for the particular AI use you are considering.

The absence of a clearly labelled AI policy therefore creates uncertainty, not automatic permission.

Policy Silence Is Not the Same as Permission

A Missing AI Page Does Not Mean There Are No Applicable Rules

Check the Publisher Before Concluding That No Policy Exists

Many journals operate under publisher-wide editorial policies. The journal's own author instructions may therefore contain little or no AI-specific wording because the relevant requirements are maintained centrally.

Elsevier, for example, maintains publisher-level generative AI policies for journal authors, reviewers, and editors. Its current author policy addresses manuscript preparation, human accountability, disclosure, privacy and confidentiality, authorship, and AI used as part of the research process. Springer Nature similarly maintains publisher-level editorial policies covering AI-assisted manuscript preparation and visual content.

Consequently, searching only the individual journal's webpage can produce a false negative.

This search should precede the conclusion that the journal genuinely has no clear policy.

General Editorial Rules Still Apply to AI-Assisted Work

Even when a journal has no explicit AI policy, ordinary publication requirements do not disappear.

If the journal requires authors to take responsibility for the accuracy of submitted work, AI assistance does not remove that responsibility. Requirements concerning originality, plagiarism, accurate citation, research methods, image integrity, participant confidentiality, copyright, competing interests, and authorship may also remain applicable regardless of whether AI is mentioned.

The International Committee of Medical Journal Editors currently recommends that journals maintain an AI policy and make authors, reviewers, and editors aware of it. ICMJE also establishes broader principles of human accountability, confidentiality, attribution, and transparency for AI use in publishing. These recommendations are influential, particularly in biomedical publishing, but they should not be silently treated as the policy of a journal that has not adopted them.

Determine What the AI Actually Did

The appropriate response to policy silence depends partly on the nature of the AI use. "I used AI" is too broad to assess.

The more AI influences scholarly content, methods, evidence, or interpretation, the less defensible it becomes to rely casually on policy silence.

Do Not Confuse No Journal Policy With No Other Policy

A journal is only one authority in the research process. Your university may regulate the AI system, your ethics approval may constrain participant-data processing, your funder may impose AI conditions, and a data-use agreement may prohibit third-party processing.

Those requirements continue to apply even if the journal says nothing about AI. The broader policy framework governing AI-assisted research does not vanish because one layer is silent.

Similarly, knowing that journals can impose requirements beyond university permission does not imply the reverse proposition that journal silence overrides university restrictions.

Consider Whether the AI Use Belongs in the Methods Section Anyway

Some AI use should be reported because it is methodologically relevant, not merely because a publisher has created a special AI disclosure rule.

If an AI system classified data, generated analytical code, processed images, transcribed recordings, performed model-based analysis, or otherwise materially affected how the study was conducted, readers may need that information to understand, evaluate, or reproduce the research.

Elsevier's current policy illustrates the distinction by directing authors to describe AI used in the research process in the Methods section, separately from its declaration requirements for AI used during manuscript preparation.

A journal's silence on a dedicated "AI declaration" therefore should not be interpreted as permission to omit methodologically necessary information.

Transparency Is Useful, but Do Not Invent a Required Declaration Format

When substantive AI use is relevant to understanding the work, transparency is generally prudent. Yet there is a difference between transparently reporting what you did and falsely claiming that a journal requires a particular declaration.

If the journal provides no prescribed wording or location, avoid copying another publisher's mandatory declaration verbatim and presenting it as though the target journal required it. Instead, report relevant AI use in the manuscript where it naturally belongs or seek editorial guidance when the placement matters.

Contact the Journal When the Answer Could Affect Submission

Not every missing sentence requires an email to the editor. A journal can reasonably become weary of receiving philosophical inquiries about whether a spellchecker counts as artificial intelligence. Researchers have enough correspondence already.

Clarification becomes more useful when AI contributed substantially to manuscript content, generated or modified visual material, was used in a way that might affect authorship or originality expectations, or created uncertainty about a submission declaration.

A concise query should describe the actual use rather than asking, "Do you allow AI?" For example, identify the type of tool, what it did, whether it affected scholarly content, and what you intend to disclose.

Document What You Did While the Policy Is Unclear

Keep enough information to reconstruct consequential AI use: the tool or service, purpose, relevant version where material, type of input, nature of the output, extent of human review, and how the output entered the research or manuscript.

This documentation gives you options. If the journal later asks for a declaration, an editor requests clarification, or the policy changes before acceptance, you are not forced to reconstruct the workflow from memory.

Check Again Immediately Before Submission

AI publishing policies are changing quickly. Elsevier, for example, updated its generative AI journal policies in 2026, while Springer Nature currently applies a risk-based framework to AI-assisted manuscript preparation. ICMJE substantially expanded its AI recommendations in its January 2026 update.

A policy absent when you began drafting may therefore exist by the time you submit. Researchers should track relevant policy changes during longer projects rather than relying indefinitely on an earlier search.

Submitting After Substantive AI-Assisted Editing

Common Mistakes When a Journal Is Silent About AI

Match Your Response to the Consequence of the AI Use

If editorial guidance creates an incompatibility with institutional requirements, move from policy silence to the separate problem of resolving a university-journal AI policy conflict.

Before Submitting to a Journal With No Obvious AI Policy

Frequently Asked Questions When Journals Have No Clear AI Policy

Silence Creates Uncertainty, Not a Blank Cheque

How to Cite This Guide

This guide is intended to be read, shared, and used in research, teaching, and academic work. If you draw on its ideas, explanations, or other content, please acknowledge the source by citing the guide. Doing so gives appropriate credit and helps your readers locate the original resource.