AI disclosure statement

AI Disclosure Statement for Research Papers: What Authors Should Write in 2026

Mikky Publication ExpertsJuly 22, 202620 min read
Research paper with AI disclosure and academic publishing documents on a professional workspace.

AI Disclosure Statement for Research Papers: What Authors Should Write in 2026

AI is now a standard feature of research and publishing in the academic world. Generative AI tools are being applied in a wide variety of ways, such as enhancing grammar and translation, generating ideas, summarizing literature, writing code, and aiding in data analysis.

As AI becomes more prevalent in the research landscape, however, there is one question that continually arises:

"If I used AI while preparing my research paper, do I need to disclose it?"

It depends on how you used the tool, what impact it made on your work and on the journal's policy.

This is where an AI disclosure statement is important.

An AI disclosure statement is a brief statement informing editors, reviewers, and readers about the use of AI in the research or manuscript writing process and, if applicable, how. It's not an attempt to make authors feel guilty about technology. It's about being honest.

There is no one rule of disclosure for all the journals, however. The AI disclosure policy journals vary widely and authors should verify the disclosure requirements of the paper's target journal before submission.

Let's explore what an AI disclosure statement in research papers should include, when to mention the use of AI, where to place the statement, and how to craft the AI disclosure statement in 2026.

What Is an AI Disclosure Statement?

AI disclosure statement: Declaration by author(s) disclosing the use of artificial intelligence in the research, writing, editing, analysis or other aspects of the publication process.

An author could use a generative AI tool to, for instance:

  • Improve grammar and sentence structure
  • Translate content from one language to another
  • Brainstorm research ideas
  • Create an outline
  • Summarize research papers
  • Help organize a literature review
  • Draft or rewrite parts of a manuscript
  • Generate or debug computer code
  • Assist with data analysis
  • Create figures or images
  • Support other research-related tasks

Disclosure is not always required for all types of AI usage. Spelling or grammar support is more likely to be classified as a different service from providing large passages of a manuscript or part of a research analysis via AI.

In this regard, it is important for an effective disclosure using generative AI to be specific enough that it explains what's happened but doesn't cause confusion.

There are usually 4 questions which should be addressed by a simple disclosure:

  • Which AI tool was used?
  • Why was it used?
  • What part of the work did it assist with?
  • Did the authors review and verify the output?

The last point is particularly important. Authors remain responsible for the accuracy, originality, integrity, and final content of their research paper.

Why Is AI Disclosure Important in Academic Publishing?

While AI tools can be helpful, they also pose potential threats to scholarly communication.

Generative AI systems may sometimes generate:

  • Incorrect facts
  • Invented references
  • Misleading summaries
  • Biased information
  • Incorrect interpretations
  • Fabricated quotations
  • Inaccurate statistical explanations

Transparency is important, therefore.

AI disclosure statement for research papers assists readers to comprehend the part AI performed in the research. It also provides editors and reviewers with a better understanding of the research and manuscript development process.

Most importantly, there is no admission of wrongdoing when disclosing.

Responsible use of an AI tool is quite distinct from using AI tools to fabricate research data or manipulate data results or to submit a generated article without proper verification.

The basic principle is simple:

Always use AI responsibly, acknowledge any use of AI that may affect the final product and be accountable for the final product.

Do You Need to Disclose AI Use in a Research Paper?

This is likely to be the greatest question for authors.

The answer is not just 'yes,' because AI can have numerous applications.

It's helpful to conceptualize AI use in three general categories.

1. AI Use That May Require Disclosure

Generally, the authors should think about disclosure if AI has contributed meaningfully and substantively to the research or the manuscript.

Examples include:

  • Generating substantial manuscript text
  • Rewriting significant sections of a paper
  • Summarizing literature that influenced the research
  • Assisting with research analysis
  • Generating or modifying research-related code
  • Producing figures or images
  • Generating synthetic research data
  • Assisting with interpretation of research findings
  • Contributing to methodological decisions

In such cases, it is generally best to be clear about the disclosure, which should be consistent with the journal's policy.

2. AI Use That May Depend on Journal Policy

There are some uses that are in a grey area.

For example:

  • Grammar correction
  • Spelling correction
  • Language polishing
  • Translation
  • Improving readability
  • Basic formatting assistance

Some journals will not require authors to disclose simple language assistance, and some will require authors' disclosure of any use of generative AI.

That is why authors are advised to always read the journal policy before submitting.

When instructions are not clear it is better to ask the journal editorial office rather than take a chance.

3. AI Use That Requires Special Consideration

The applications of AI go beyond manuscript editing in some cases.

For example:

  • Generating research data
  • Creating experimental results
  • Producing images that represent research findings
  • Making decisions that affect study methodology
  • Conducting substantive data analysis
  • Generating code used for important statistical procedures

Some of these uses may necessitate specific disclosure and can also lead to concerns about research integrity, research reproducibility, and methodology.

A brief statement about the use of AI in the preparation of this paper may not be sufficient.

Authors should describe the function of the tool clearly and adhere to the requirements of the journal they are submitting to.

A Simple AI Disclosure Decision Framework

Before you turn in your paper, you should ask yourself:

Did AI only help with basic language correction?

If so, review the policy of the journal. There is no requirement for disclosure, but it may be disclosed.

Did AI substantially rewrite or generate manuscript content?

If so, disclosure is typically recommended, and may be required by the journal's policy.

Did AI influence your research process?

Consider a more detailed disclosure if AI was used in some aspect of literature analysis, research design, coding, data analysis, or data interpretation.

Did AI generate data, figures, or images?

If so, read the journal's policy and give detailed information on the tool used and how it was used.

Did AI produce something you included without verification?

Stop and check it before submitting. It is the author's responsibility to verify content created with AI.

AI Disclosure Rules for Authors in 2026

At the moment there's no universal AI disclosure policy that the global community of journals can adhere to.

Publishers and scholarly organisations have their own strategies and the needs may evolve with advances in AI.

Thus, authors are advised to verify:

  • The journal's author guidelines
  • The publisher's AI policy
  • The journal's peer-review instructions
  • Institutional research policies
  • Funding agency requirements
  • Field-specific ethical guidelines

What matters is that these rules on the disclosure of the use of AI can vary between journals.

Some journals, for instance, may require authors to report on the use of generative artificial intelligence tools for manuscript preparation, whereas others may differentiate between simple editing and generative-AI generated content.

That's why it is not always a good idea to copy an AI statement from another paper.

Your statement should effectively articulate what you have used AI for and meet the requirements of the journal to which you are submitting.

What Do Major Publishers Say About AI Disclosure?

Authors often search for specific requirements from major publishers before submitting their papers. While policies can change, the following publisher and organization guidance is particularly relevant when preparing a manuscript in 2026.

Elsevier Generative AI Policy for Authors

The Elsevier generative AI policy for authors primarily encourages transparency with the use of generative AI tools to prepare manuscripts.

Authors should distinguish between the use of AI for the purpose of improving the ordinary language of the text and more substantive generative AI assistance. In cases where disclosure is mandated, the authors should explain the use of the technology appropriately.

It is also important that writers be aware that generative AI is not responsible for the manuscript. The final product is the responsibility of the human writers.

Always make sure you are submitting a manuscript to the most up-to-date Elsevier journal submission requirements as each journal has its own requirements.

Springer Nature AI Authorship Policy

It's critical to note that the Springer Nature AI authorship policy differentiates between the use of AI tools and authorship.

AI tools cannot be credited as authors as they cannot be held responsible for the research, approve the final manuscript, or answer questions regarding the research.

It is the responsibility of human authors to remain responsible for the manuscript.

In addition, authors are required to adhere to the current Springer Nature requirements in reporting use of artificial intelligence to assist with the writing and content of their article and/or other material content.

Taylor & Francis AI Disclosure Requirements

Transparency and responsible use are the key considerations in the Taylor and Francis AI disclosure requirements.

Please refer to the most recent instructions for authors from Taylor & Francis before submission, particularly if you have used generative A.I. to generate or significantly alter the content of your manuscript.

It is important that authors don't take it for granted that the following disclosure does not apply to every Taylor & Francis journal when using AI tools for writing. Please, verify the requirements of the particular publication.

SAGE AI Disclosure Policy

The SAGE AI disclosure policy also speaks to the growing trend in scholarly communication for disclosure.

When preparing manuscripts and/or conducting research with AI tools, authors should consult the most recent guidance on the usage of AI tools available in the journal.

Authors should describe the role of AI in a substantive manner if it has made a substantive contribution, rather than a generic statement.

ICMJE AI Disclosure Guidelines

The ICMJE AI disclosure guidelines are particularly relevant for authors publishing biomedical and medical research.

The ICMJE has emphasized transparency around the use of AI and the continuing responsibility of human authors for the accuracy and integrity of submitted work.

Authors working in medical research should pay close attention to ICMJE recommendations as well as the specific instructions of the journal they are targeting.

IEEE AI Disclosure Policy for Authors

The IEEE AI disclosure policy authors should follow may vary according to the type of AI use and the publication involved.

Researchers submitting to IEEE journals and conferences should carefully review the current publication-specific requirements.

AI-assisted writing, AI-generated content, and AI-generated images may be treated differently from basic grammar or spelling assistance.

The safest approach is to identify exactly how AI was used and then compare that use with the current IEEE policy before submission.

What Should an AI Disclosure Statement Include?

You don't need to make a disclosure complicated.

A helpful formula to use is:

AI Tool + Purpose + Scope + Human Verification

For example:

"The authors used ChatGPT to assist with language editing and improve the clarity of selected sections of the manuscript. All AI-assisted text was reviewed and revised by the authors, who take full responsibility for the final content."

In the statements below, the reader is told:

  • The tool used
  • The reason it was used
  • The scope of assistance
  • The fact that humans reviewed the final output

Other technical details, such as the mention of the AI model, version, date of use, etc., may be required depending on the journal's requirements.

Where Should You Put an AI Disclosure Statement?

No single location is suitable for all journals.

An AI disclosure statement could be included in the following locations, depending on the journal's instructions:

  • A dedicated AI disclosure section
  • Declarations
  • Acknowledgments
  • Methods
  • Author contributions
  • Supplementary information

The Methods section might be applicable, depending on whether AI was a component of the research.

If AI-assisted coding was incorporated into a data analysis process, for instance, the Methods section of your paper could benefit from a description of the contributions of the AI tool.

If only the AI has been utilized in the preparation of the manuscript, the journal may prefer a declaration or acknowledgments including a statement of how the AI was used.

Be sure to follow the directions given on the journal.

AI Disclosure Statement Examples for Research Papers

The list of examples below is available for the authors to adapt. They should not be copied verbatim as the wording should be a true reflection of the use of AI.

1. AI Used for Grammar and Language Editing

The authors employed [Tool Name] to help them with grammar, spelling and language editing in preparing a manuscript and accept full responsibility for what has been written.

2. AI Used for Brainstorming

The authors employed [Tool Name] in the early stages of developing the manuscript to brainstorm and organize ideas. The research questions, analyses, interpretations and conclusions were developed by the authors independently and verified.

3. AI Used for Literature Summarization

To support the summarizing of selected literature, [Tool Name] was used by the authors in the initial research process and all relevant sources were reviewed and verified independently by the authors.

4. AI Used for Manuscript Drafting

The authors have utilized [Tool Name] during the process of writing selected portions of this manuscript, and have carefully reviewed, edited and verified all added output of the tool, for which the authors accept full responsibility for the final product.

5. AI Used for Translation

"The authors utilized [Tool Name] for the purposes of translating while preparing the manuscript, and then reviewed and corrected it for accurate and meaning.

6. AI Used for Coding

The authors have utilized [Tool Name] for developing and testing a part of the computational code employed in this study, which was then reviewed and tested by the authors and for which the authors take responsibility for the accuracy of the analysis.

7. AI Used for Data Analysis

The author(s) have used [Tool Name] to support some aspects of data analysis and the analysis procedures and results were independently reviewed and validated by the author(s), who is responsible for the interpretation and reporting of the results.

8. AI Used to Create Figures or Images

"The authors used [Tool Name] to assist in creating [describe figure/image]. The generated material was reviewed and modified by the authors and is disclosed in accordance with the journal's requirements."

AI Disclosure Matrix: What Should You Disclose?

AI UseDisclosure ApproachWhat to Mention
Grammar and spellingCheck journal policyTool and purpose if required
Language polishingOften policy-dependentTool and scope
BrainstormingPolicy-dependentHow AI supported ideation
Literature summarizationConsider disclosureTool and role
Literature discoveryDepends on impactHow AI influenced source selection
Manuscript draftingGenerally discloseTool and sections assisted
TranslationCheck journal policyTool and language assistance
CodingDisclose when materialTool and coding contribution
Data analysisDiscloseTool and analytical role
AI-generated figuresDiscloseTool and figure creation
AI-generated dataRequires special attentionTool, method, and data role
Reference generationVerify carefullyTool use and reference checking

The key idea is that the more AI influences the actual research or published content, the more important a clear disclosure becomes.

AI Disclosure Is Not Just About Writing

One common mistake is to think that AI disclosure only applies when ChatGPT or another tool writes sentences.

It's not sufficient anymore.

By 2026, researchers can use AI throughout the research workflow.

For instance, AI can assist with:

Research Design

Brainstorm research questions or research approaches using an AI tool.

Literature Review

AI can be used to search, categorize, or synthesize pertinent literature.

Data Analysis

AI tools can help researchers analyze data or create analysis methods. AI tools can help researchers interpret data or develop analysis methods.

Programming and Coding

Generative AI tools can be used for coding, code explanations, or code debugging.

Figures and Images

Illustrations or other visual content may be generated with the help of AI.

Translation

AI-powered translation is useful for researchers involved in multiple languages.

Manuscript Preparation

AI can help you draft and edit, rewrite or improve clarity.

The disclosure required may vary as per the use and requirements of the journal.

AI Disclosure vs AI Authorship: What Is the Difference?

This is a place where it is easy for many new researchers to get confused.

Disclosure is not necessarily authorship.

An AI tool cannot not be a co-author because it cannot:

  • Accept responsibility for the research
  • Approve the final manuscript
  • Respond to editorial questions
  • Take responsibility for research integrity
  • Declare conflicts of interest
  • Give consent to publication

The research paper is the responsibility of the human authors.

As such, the use of ChatGPT, Gemini, Claude or any other generative AI tool during manuscript preparation should NOT automatically make it part of the author list.

Rather, comply with the disclosure requirements of the journal.

What Authors Should Not Write in an AI Disclosure

A disclosure should be straightforward, precise and truthful.

Don't make sweeping generalizations.

Weak:

This research involved the use of AI.

This leaves even more questions unanswered.

What AI tool?

For what purpose?

Was it used for writing, analysis, coding or other?

Better:

"The authors used ChatGPT to assist with language editing and improve the clarity of selected sections of the manuscript. All AI-assisted content was reviewed and revised by the authors."

It follows with the second statement which gives useful information without over-explaining.

It's also important to authors that they do not assert that AI was not used when in fact it was used in an AI disclosure method.

It is best to be transparent rather than to make it seem insignificant to use AI.

Bad vs Good AI Disclosure Statements

Bad StatementBetter Statement
"AI was used to write this paper.""ChatGPT was used to assist with language editing and improve clarity in selected sections."
"The paper was prepared using AI.""The authors used [Tool] during manuscript organization and brainstorming."
"AI helped with research.""The authors used [Tool] to summarize selected literature during the initial research stage; all sources were independently reviewed."
"The authors used ChatGPT.""The authors used ChatGPT to assist with grammar and language editing. All output was reviewed and revised by the authors."

What If You Forgot to Include an AI Disclosure?

It happens.

Or maybe you did an AI assist while writing your manuscript and you found out the journal you're sending it to requires disclosure. Or perhaps you used an AI tool for manuscript preparation and realized disclosure is required by the journal you're sending it to.

Don't attempt to sweep the problem under the rug.

Instead:

Step 1: Review the Journal's Policy

Please refer to the most up-to-date instructions for authors.

Step 2: Identify How AI Was Used

Record precisely what the tool did.

Step 3: Determine Whether Disclosure Is Required

Compare usage to journal's policy.

Step 4: Add the Appropriate Statement

Update manuscript if it hasn't been submitted.

Step 5: Contact the Editorial Office if Necessary

If the paper was submitted, see if the journal will accept the disclosure to be added or corrected.

The last thing you want to do is to cover up relevant AI use just because you're afraid it will be viewed negatively.

Will an AI Disclosure Statement Cause Your Paper to Be Rejected?

One of the top worries of researchers is this.

It depends on which journal, type of AI use and if it complies with the journal policy.

Obviously, a disclosure alone does not necessarily point to poor research.

The important difference between:

  • Responsible AI assistance
  • Undisclosed substantive AI use
  • Fabricating research data
  • Generating false references
  • Misrepresenting AI-generated content as entirely human-created

The real issue is whether the research remains trustworthy and whether the authors have followed the relevant rules.

When using AI tools responsibly, ensuring the content is accurate and adheres to the guidelines provided by the journal, transparency is usually preferable over obscurity.

AI Disclosure and the Problem of AI Hallucinations

AI tools are able to sound confident even when they are incorrect.

This is of great risk especially in academic research.

Prior to submission of an article that uses AI to aid in writing, authors should check:

  • Every citation
  • Every reference
  • Statistical claims
  • Quotations
  • Facts and figures
  • Equations
  • Research findings
  • Methodological descriptions
  • Data interpretations

Do not trust any reference from AI, even if it appears to be from an academic source.

An AI disclosure statement isn't a shifting of liability from the author to the AI tool.

The author will be responsible for the final research paper.

2026 AI Disclosure Checklist for Authors

Before you send your manuscript written with the aid of AI, consider the following questions:

  • Have I checked the target journal's AI policy?
  • Do I know exactly how AI was used?
  • Did AI generate or substantially modify manuscript content?
  • Did AI assist with data analysis or coding?
  • Did AI generate any figures or images?
  • Did AI influence my literature review?
  • Have I verified every AI-generated citation?
  • Have I checked all factual claims?
  • Have I reviewed the AI-generated content myself?
  • Have I written an accurate AI disclosure statement?
  • Did I place the statement where the journal requires it?
  • Have I avoided listing AI as an author?
  • Am I taking full responsibility for the final manuscript?

The more you can answer these questions, the better you'll be able to submit your paper responsibly.

A Simple Formula for Writing Your Own AI Disclosure Statement

When you're not sure how to begin, follow this format:

"The authors used [AI TOOL] for [PURPOSE] during [STAGE/AREA OF WORK]. The AI-generated or AI-assisted output was [REVIEWED/EDITED/VERIFIED] by the authors, who take full responsibility for the final content."

For example:

All the text prepared with ChatGPT was reviewed, edited and verified by the authors, who are responsible for the accuracy and integrity of the final product.

It's a great starting point but keep in mind that the actual wording of your AI use and the requirements of your target journal should be reflected in your final language.


Frequently Asked Questions

1. What is an AI disclosure statement?

An AI disclosure statement is used to report the use of an AI tool in the research or manuscript preparation process. The tool and its purpose, scope and human verification process may be described, depending on the journal's policy.

2. Do I need an AI disclosure statement for every research paper?

Not necessarily. It depends on the use of AI and the journal where you are submitting. Please always refer to the current author guidelines for submission.

3. Should I disclose ChatGPT if I only used it for grammar correction?

It will be determined by the journal's policies. Certain journals differentiate between language assistance and substantive generative AI use, and others might have more expansive disclosure demands.

4. Can ChatGPT be listed as a co-author?

Typically, the answer is No: AI cannot be responsible for the research, approve the final manuscript, or answer questions about the manuscript. The publication is the responsibility of human authors.

5. Where should I put my AI disclosure statement?

Depending on the journal, the location is different. May be included in a section of the document designated for AI disclosure, declarations, acknowledgments, Methods, or other specific sections.

6. Do I need to disclose AI used for literature searching?

The extent and purpose of the use is the determining factor. Disclosure may be warranted if AI substantially contributed to the identification, selection, summarisation or interpretation of literature. It is also the responsibility of the author to check his/her own sources.

7. Will disclosing AI use cause my research paper to be rejected?

Not automatically. The key ones are: journal policy; the type of use of AI; research integrity; and whether the authors have reported and verified the work properly.


M

Mikky Publication Experts

Mikky Publication Editorial

Published July 22, 2026

Ready to Publish?

Share Your Research
With the World.

Get expert guidance from submission to publication. Let your research reach readers worldwide.