Generative AI Tools

IOP Publishing is committed to supporting innovation in research and scholarly communication. As generative AI is increasingly used across the research ecosystem, we support its responsible use in ways that enhance discovery, accessibility and research quality while upholding the highest standards of integrity and transparency. This policy outlines acceptable and unacceptable uses of generative AI (GenAI), including large language models (LLMs) and AI chatbots, by authors in IOP Publishing journals.

Why this policy matters

There are many responsible and appropriate uses for GenAI within scholarly research and we support authors using it in this manner. For example, they may help authors as part of the research process, overcome language barriers or more efficiently process data. However, these tools can produce misleading or fabricated content, cannot be legally accountable for published work and lack the ability to think critically about the material they are producing.

Additionally, uploading manuscript material to GenAI platforms may expose sensitive data to third parties, potentially breaching the rights of others involved in the work, including authors, participants, data owners and reviewers.

This policy aims to balance these considerations, safeguarding confidentiality, accuracy, and fairness while allowing transparent use of AI by authors. As community expectations and industry standards for AI disclosure evolve, IOP Publishing will continue to review and update its requirements to support responsible innovation, research integrity, transparency, and trust in the scholarly record.

Use of Generative AI tools for research purposes

GenAI tools may be used as part of the research process, including to support research ideation, methodology, analysis where their use is methodologically appropriate, transparent and reproducible. All authors remain fully responsible for the accuracy, originality and integrity of all content in their submission, including any content, analysis or conclusions produced with the support of AI tools.

Acceptable Uses

Authors may use generative AI tools to:

Assist with the research process – including hypothesis generation, research ideation, experimental design and optimisation, simulation and model selection, code generation, data analysis and anomaly detection, and the conduct of computational or AI-enabled research workflows, provided their use is disclosed where required and any outputs are appropriately validated by the authors.
Edit human-written text – this includes minor corrections (checking spelling, grammar and punctuation) and more significant changes (enhancing the clarity and structure of the work).
Generate text – authors must ensure that they have critically revised any AI-generated text to ensure it is accurate and free from plagiarism. Authors are responsible for the interpretation, reasoning and conclusions presented in the manuscript.
Generate figures based on existing data – for example, creating a graph based on data collected during an experiment. This includes the use of AI-assisted coding, visualisation and formatting tools however authors must ensure that AI-assisted figure generation does not alter, distort or misrepresent the underlying data.
Create explanatory images – such as graphical summaries, conceptual illustrations, workflow diagrams, flowcharts and decision trees, where these are used to communicate research concepts, methods, or processes rather than present research data or results. Any AI-generated explanatory image must be clearly distinguishable from research data and must not misrepresent, fabricate, or obscure scientific evidence.
Support literature review – GenAI tools may be used to help locate, retrieve, organise and summarise relevant publications for authors to read and draw upon in their own manuscript. However, authors are responsible for the interpretation, synthesis and critical evaluation of the literature, and literature review sections should reflect the authors’ own scholarly assessment of the field.
Edit their responses to peer-review reports – authors may use GenAI tools only to improve the language of their responses to peer review reports.

Disclosure

Clear and transparent disclosure of the use of AI tools supports trust, reproducibility, and the integrity of the scientific record. Where AI tools have been used in ways that materially contribute to the research process, the preparation of the manuscript or the analysis, interpretation and presentation of results, authors must disclose this in a separate AI disclosure statement as part of the Acknowledgement section of their article. This ensures that its role is visible to readers, reviewers and downstream users.

The disclosure statement should identify the tool or model used, describe how it was used and explain the extent to which the authors verified, validated and critically reviewed the AI-assisted outputs. Where AI tools have been used as part of the research process rather than manuscript preparation, this use should be additionally described in detail in the Methods (and other relevant sections) of the main body of the article.

Template disclosure statement:
“During the preparation of this manuscript, the authors used [Tool Name, Version] to support [describe use]. The authors reviewed, validated and critically assessed all AI-assisted outputs and take full responsibility for the content of the manuscript.”

PLEASE NOTE: Where AI tools have been used for routine spelling, grammar and language editing that does not materially contribute to the manuscript a separate disclosure statement is not required.

Authors are encouraged, and may be required in fields where AI is central to the research method, to provide relevant prompts, prompt logs, model settings, outputs and validation materials as supplementary data where this is necessary to support transparency, reproducibility and assessment of the work. This requirement will be reviewed by IOP Publishing as community standards for AI-use disclosure and metadata develop.

All named authors are fully responsible for all material presented in their manuscript, and for ensuring its accuracy, integrity and originality.

Unacceptable Uses

Authors may not:

Fabricate original research data or results – any data or results must have been gathered from the experiment presented in the paper or generated through a legitimate, disclosed and scientifically valid research method.
Alter or manipulate original research data or results – this includes the manipulation of images such as blots.
Generate or modify reference lists – authors may use generative AI tools to support literature discovery, citation management, reference formatting and reference checking, but every cited source must be verified by the authors against the original source to confirm that it exists, is accurately represented, and is drawn upon and referenced in the main body of the manuscript.
Upload reports from reviewers to generative AI tools – this may expose sensitive data to third parties, breaching reviewer rights and privacy laws.
Generate responses to reviewers – it is important that authors critically engage with reviews and revise their work based on the advice of their peers. Authors must personally evaluate reviewer comments, verify any AI-assisted outputs and take full responsibility for the final response.
Hidden prompts – concealing prompts for GenAI tools within a manuscript in an attempt to manipulate the peer review process will result in immediate rejection.

Literature review

While GenAI tools can be used to support literature review, authors should keep in mind that these tools are prone to generating false content, including references to non-existent work.

GenAI tools may assist in identifying, retrieving, organising and summarising relevant literature. However, authors must independently assess, interpret and synthesise the literature and ensure that any conclusions or characterisations of prior work accurately reflect the cited sources.

We consider the presence of references to non-existent sources to be strong evidence of irresponsible AI usage and to raise serious concerns about the validity of the work. If they are found during the submission process, this will usually result in a rejection of the submitted manuscript and potentially further sanctions. If they are discovered after publication (either in the Accepted Manuscript or Version of Record), IOP Publishing reserves the right to retract the paper due to a loss of confidence in the work.

Administrative mistakes where references to two existing papers have been mixed-up, or where there is a typo in a reference to an existing paper, are not covered by this policy.