Can Generative AI Help Draft a Data Availability Statement?

Can AI Write the Data Availability Statement Required by Your Journal?

Your manuscript is ready for submission, but the journal requires a data availability statement. You need to explain whether the data supporting your findings are publicly accessible, available upon request, restricted, or unavailable for sharing.

You might ask generative AI to draft the statement. After all, many journals use familiar wording for data availability declarations, and AI can produce a polished paragraph within seconds.

The difficulty is that a data availability statement is not merely administrative text. It makes factual claims about the existence, location, accessibility, and conditions of use of research data.

If AI generates a statement that does not reflect the actual data-sharing arrangements, the manuscript may mislead editors, reviewers, and future researchers. The question is therefore not simply whether AI can write the statement, but whether the author can substantiate everything it says.

Yes, but AI Must Work From Verified Data-Sharing Information

What AI Needs to Know Before Drafting a Data Availability Statement

What Is a Data Availability Statement?

A data availability statement explains how readers can access the research data supporting the findings reported in a publication. Where access is limited, it should describe the relevant restrictions and, when appropriate, the process through which access may be requested.

Depending on the journal and research design, the statement may identify a repository, a persistent dataset identifier, supplementary materials, a controlled-access process, or a justified reason why data cannot be shared.

Its purpose is to communicate the actual availability of the underlying research data. It does not, by itself, establish that the research is reproducible or that the data are sufficiently documented for reuse.

Springer Nature and Wiley publish guidance and example statements for different data availability situations. Their approaches illustrate why authors should follow the target journal's requirements rather than assume that one generic statement is acceptable everywhere.

What Information Must Be Verified Before Using AI?

Before generating the statement, establish which datasets support the reported findings and what access arrangements actually apply.

Useful information includes the dataset's location, repository name, persistent identifier where available, access restrictions, responsible data custodian, and any conditions governing reuse.

Authors should also determine whether the relevant data are owned or controlled by their institution, research collaborators, a third-party provider, or another organization.

AI should receive confirmed facts rather than being asked to infer the sharing arrangements from the manuscript's methods section.

Different Data Availability Situations Require Different Statements

The appropriate wording depends on how the data can actually be accessed.

These categories are practical distinctions rather than universal journal-approved templates. Some studies involve several datasets with different access conditions and require a statement that describes each arrangement separately.

Can AI Draft a Statement for Publicly Available Data?

Yes, provided the author supplies a verified repository record and confirms that the supporting data are accessible under the stated conditions.

For example, a statement may identify a dataset deposited in Zenodo, Dryad, or another suitable repository, together with its persistent identifier.

However, uploading files to a repository does not necessarily mean they are publicly available. A deposit may remain private, embargoed, incomplete, or awaiting publication.

AI should not claim that a dataset is openly accessible merely because a repository name appears in the author's notes.

Before submission, the author should check that the identifier resolves correctly and that the repository record corresponds to the dataset used in the manuscript.

Is a DOI Enough to Establish Data Availability?

A digital object identifier can provide a persistent reference to a dataset record, but it does not establish that every underlying file is openly accessible.

The record may describe restricted data, an embargoed deposit, or material requiring authorization.

Consequently, the statement should distinguish between the availability of dataset metadata and the accessibility of the data themselves.

AI can format a verified DOI into a statement, but it should not invent identifiers or infer access permissions from their existence.

Can AI Write That Data Are Available Upon Reasonable Request?

AI can draft such wording, but the declaration must reflect a genuine access arrangement.

The familiar phrase "available from the corresponding author upon reasonable request" should not be used automatically. It may be unsuitable if the corresponding author does not control the data, institutional approval is required, or the relevant journal expects repository deposition.

Some publishers and funders also have policies that favor more direct data accessibility or require a justification for restricted sharing.

Authors should establish who can authorize access, what conditions apply, and whether requests can realistically be fulfilled.

What If Research Data Contain Confidential Participant Information?

Research involving interviews, medical records, student information, or other sensitive material may be subject to ethical and legal restrictions.

Removing names does not necessarily make a dataset anonymous. Participants may remain identifiable through combinations of demographic characteristics, contextual details, or distinctive quotations.

A data availability statement should therefore reflect the consent arrangements, ethics approval conditions, institutional policies, and applicable data protection requirements.

For research conducted in the Philippines, the Data Privacy Act of 2012 and relevant National Privacy Commission guidance may be applicable, depending on the nature of the personal data and the processing activities.

Authors should not assume that a journal's preference for data sharing overrides participant confidentiality or applicable law.

What If the Data Cannot Be Shared at All?

Some research data cannot be made available because of consent restrictions, confidentiality obligations, contractual agreements, or other legitimate constraints.

In such cases, the statement should describe the restriction accurately and with enough specificity to explain why access is unavailable.

For example, a qualitative study may involve interview transcripts containing information that cannot be sufficiently de-identified for unrestricted release.

Whether the journal accepts that arrangement depends on its policies and the circumstances of the research. Authors should not fabricate an access route simply to make the statement appear compliant.

Where permitted, alternative materials such as coding frameworks, aggregate results, or appropriately de-identified excerpts may support transparency without exposing restricted records. Their availability should be described accurately rather than presented as equivalent to access to the complete dataset.

How Does a Data Availability Statement Relate to FAIR Data?

The FAIR principles encourage research data to be Findable, Accessible, Interoperable, and Reusable.

Importantly, FAIR does not mean that every dataset must be openly accessible without restriction. The principles recognize that access may require authentication or authorization, provided the access conditions and metadata are appropriately described.

A data availability statement can help communicate these arrangements, but a well-written statement alone does not make a dataset FAIR.

Achieving meaningful reuse may also require adequate metadata, documentation, licensing information, and suitable data formats.

Do Journals Require Different Data Availability Statements?

Yes. Journal policies vary in their expectations concerning repository deposition, supporting data, persistent identifiers, restricted access, and statement wording.

Some journals provide standard templates. Others permit free-text explanations or require specific details within the manuscript submission system.

Authors should use the journal's current official instructions when checking manuscript requirements. AI-generated interpretations of journal policy may be outdated or incomplete.

When journal requirements conflict with legal, ethical, or contractual restrictions, the author should seek clarification from the appropriate journal or institutional authority rather than make an unsupported declaration.

Drafting a Data Availability Statement for Confidential Interview Data

This wording is appropriate only for the hypothetical circumstances described. A real study may require additional information about institutional access procedures, supporting materials, or the precise restrictions imposed by the ethics approval.

Notice that the statement does not automatically promise access upon request. Such a promise would be misleading unless an actual authorized access procedure existed.

Common Mistakes in AI-Generated Data Availability Statements

Verify the Data Arrangements Before Asking AI to Describe Them

The safest approach is to establish the data-sharing facts independently and use AI only to express those facts clearly.

Start by identifying the datasets that support the findings. Determine whether each dataset is public, restricted, available through an approved request procedure, or unavailable for sharing. Then verify the applicable journal requirements.

Where the situation is uncertain, resolve it with the appropriate data custodian or institutional authority before finalizing the statement.

A Reusable AI Prompt for Data Availability Statements

Once the statement has been drafted, compare it with the actual repository records, consent conditions, and journal instructions. The final check should be part of verifying AI-assisted journal submission materials.

A clear statement is valuable only when it accurately describes the research data and the conditions under which other researchers may access them.

Data Availability Statement Verification Checklist

Frequently Asked Questions About AI-Assisted Data Availability Statements

AI Can Describe Data Availability, but It Cannot Establish It

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.