The DOI matches its Crossref record; 15 of the 15 references it cites were found in a registry or an index; none of the 14 sources its claims cite is checked for support yet.
A call for clarity: a unified checklist for reporting use of large language models in writing scientific manuscripts
Open accessAbstract
The rapid integration of generative artificial intelligence (Gen AI) into academic writing has outpaced the establishment of consistent norms for responsible and transparent disclosure. Leading organizations including the International Committee of Medical Journal Editors (ICMJE), the Committee on Publication Ethics (COPE), the World Association of Medical Editors (WAME), and the European Commission, Directorate-General for Research and Innovation have issued guidance affirming that AI tools cannot be listed as authors and must be transparently disclosed. However, what remains missing is both a cross-journal consensus and guidance for authors on what constitutes acceptable versus unacceptable use of Gen AI. Furthermore, as authors may employ Gen AI at multiple, distinct stages of manuscript preparation, they currently have no standardized or granular method to report this varied use in sufficient detail. This ambiguity creates a critical gap between high-level disclosure principles and practical implementation, threatening not necessarily the integrity of the underlying research itself, but the reader's and editor's ability to objectively assess the reliability and provenance of reported findings.This paper responds to that gap by proposing a structured, domain-based framework for reporting Gen AI use in scholarly manuscripts. Drawing on a synthesis of evolving editorial statements and guidelines, we outline three domains in which Gen AI is commonly employed: conceptual contributions, linguistic assistance, and research assistance. For each domain, we distinguish uses that are generally acceptable from those that raise ethical or integrity concerns, providing examples to guide authors, reviewers, and journal editors and editorial staff.To operationalize this framework, we introduce a prototype of an online Gen AI use disclosure form that guides authors through documenting their use of Gen AI across the three domains. The tool automatically generates a standardized disclosure statement and assigns a unique, citable reference number. This reference number links to a persistent, publicly accessible summary of the declared Gen AI use, creating a transparent and auditable record. This system is proposed as a 'living' platform, designed to evolve through consensus among journal editors and editorial staff, authors, and research integrity organizations, functioning similarly to other reporting guidelines hosted by the EQUATOR Network.This system moves beyond ad hoc, narrative statements to establish a proactive and standardized disclosure process around the use of Gen AI in scholarly publishing. By embedding transparency, human accountability, and traceability directly into the publication workflow, our approach complements existing frameworks for authorship and conflict of interest. Like conflict-of-interest disclosures, AI use statements surface information that allows readers to contextualize potential risks and judge credibility for themselves. Ultimately, this work advances a practical model to strengthen trust between authors, journal editors and editorial staff, and readers, aligning the promise of generative AI with the enduring principles of research integrity.
Overview
CoChat Summary
TL;DR
The paper proposes a unified, domain-based checklist and prototype online disclosure form for reporting generative AI/LLM use in scientific manuscripts. It aims to distinguish generally acceptable from generally not acceptable uses and generate a standardized, citable disclosure statement to improve transparency and accountability.
Key findings
- Proposes three domains of LLM involvement in scholarly writing: conceptual contributions, linguistic assistance, and research assistance, each divided into “Generally Acceptable” and “Generally Not Acceptable” uses.
- States that the primary criterion distinguishing acceptable from not acceptable uses is not the task category itself but the presence and quality of human oversight, verification, and accountability.
- Introduces a working prototype interactive checklist that generates a formatted disclosure statement and assigns a unique, citable reference number linked to a persistent, publicly accessible summary of declared Gen AI use.
- Notes that ICMJE, COPE, WAME, and the European Commission have affirmed AI tools cannot be authors and must be transparently disclosed, but cross-journal consensus and granular reporting guidance remain missing.
Methods
- Conceptual/commentary paper: the framework was derived through synthesis of evolving editorial statements, institutional guidelines, and published frameworks.
- Granular tasks within each domain were iteratively grouped by nature of AI involvement (generative vs. assistive), degree of required human judgment, and potential risk to accuracy, attribution, or integrity.
- The domain structure and item taxonomy were reviewed by the authorship team for face validity; no empirical sample, data collection, or statistical analysis is reported.
Limitations
- Stated: a voluntary checklist may suffer from social desirability bias, with authors reluctant to acknowledge “generally not acceptable” uses, potentially resulting in incomplete disclosures.
- Stated: LLM-enabled literature retrieval may be constrained by licensing restrictions, paywalls, and institutional access policies, potentially affecting reproducibility and independent verification.
- Stated: guidelines will inevitably evolve, and acceptable uses may be discipline-dependent, such as differences between science/medicine and English or philosophy journals. Evident from the text: no empirical validation or pilot testing of the checklist or prototype is reported.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
0 Valid · 1 Not valid · 8 Not checked · 11 Own results and interpretations
None of the sources cited for this claim have been checked yet.
In the paper
“A meaningful limitation of any voluntary checklist is social desirability bias: authors may be reluctant to acknowledge 'generally not acceptable' uses, potentially resulting in incomplete disclosures, a concern raised by Hosseini et al. [14].”
§ A call for clarity: a unified checklist for reporting use of large language models in writing scientific manuscripts > Conclusion
What the sources say
- Not checked yetcites [14] Luo X, Tham YC, Giuffrè M, Ranisch R, Daher M, Lam K, et al. Reporting guideline for the use of Generative Artificial intelligence tools in MEdical Research: the GAMER Statement. BMJ Evid Based Med. 2025;bmjebm-2025-113825. https://doi.org/10.1136/bmjebm-2025-113825. source
Checked before — its earlier verdict is held back until a new check runs.
hedgedstates a possible cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Not checked yet
None of the sources cited for this claim have been checked yet.
In the paper
“However, the GAMER statement is limited to reporting conventions without normative guidance on which uses of LLMs are acceptable or unacceptable [14].”
§ A call for clarity: a unified checklist for reporting use of large language models in writing scientific manuscripts > Introduction
What the sources say
- Not checked yetcites [14] Luo X, Tham YC, Giuffrè M, Ranisch R, Daher M, Lam K, et al. Reporting guideline for the use of Generative Artificial intelligence tools in MEdical Research: the GAMER Statement. BMJ Evid Based Med. 2025;bmjebm-2025-113825. https://doi.org/10.1136/bmjebm-2025-113825. source
Checked before — its earlier verdict is held back until a new check runs.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Not checked yet
In the paper
“Similarly, the publication of standards for reporting qualitative research in 2014 brought coherence and rigor to an area that had often been marginalized in medical scholarship [21].”
§ A call for clarity: a unified checklist for reporting use of large language models in writing scientific manuscripts > Conclusion
What the sources say
- Not checked — no full text of the sourcecites [21] O’Brien BC, Harris IB, Beckman TJ, Reed DA, Cook DA. Standards for Reporting Qualitative Research: A Synthesis of Recommendations, Academic Medicine. 2014;89(9):1245–51. https://doi.org/10.1097/ACM.0000000000000388. source
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“In both cases, clear, consensus-driven guidelines helped transform contested practices into recognized and respected domains of health research.”
§ A call for clarity: a unified checklist for reporting use of large language models in writing scientific manuscripts > Conclusion
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
7 references still to check
A free account verifies the references that are still open.
- Probably not a substantial claim. 15
- It may state two separate claims as one. 9
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ A call for clarity: a unified checklist for reporting use of large language models in writing scientific manuscripts > Introduction
“Hosseini and colleagues argued that banning AI tools outright, even if this were considered desirable in principle, would be impractical and counterproductive because such proscription would just encourage their undisclosed use [12].”
§ A call for clarity: a unified checklist for reporting use of large language models in writing scientific manuscripts > References
“12. Lin Z. Techniques for supercharging academic writing with generative AI. Nat Biomed Eng. 2024.”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ A call for clarity: a unified checklist for reporting use of large language models in writing scientific manuscripts > Conclusion
“A meaningful limitation of any voluntary checklist is social desirability bias: authors may be reluctant to acknowledge 'generally not acceptable' uses, potentially resulting in incomplete disclosures, a concern raised by Hosseini et al. [14].”
§ A call for clarity: a unified checklist for reporting use of large language models in writing scientific manuscripts > References
“14. Luo X, Tham YC, Giuffrè M, Ranisch R, Daher M, Lam K, et al. Reporting guideline for the use of Generative Artificial intelligence tools in MEdical Research: the GAMER Statement.”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ A call for clarity: a unified checklist for reporting use of large language models in writing scientific manuscripts > Introduction
“Lissack and Meagher, meanwhile, provided a broad ethical “how-to,” emphasizing transparency, originality, verification, and responsible use, casting the LLM as a peer learner whose contributions must be guided by the researcher’s intellectual ownership [12].”
§ A call for clarity: a unified checklist for reporting use of large language models in writing scientific manuscripts > References
“12. Lin Z. Techniques for supercharging academic writing with generative AI. Nat Biomed Eng. 2024.”
Citations
Cites
von Elm, E., Altman, D. G., Egger, M., Pocock, S. J., Gøtzsche, P. C., Vandenbroucke, J. P., & Initiative, F. T. S. (2007) · Annals of Internal Medicine · doi:10.7326/0003-4819-147-8-200710160-00010 · Cited by 17,796
The PRISMA 2020 statement: an updated guideline for reporting systematic reviews
Page, M. J., McKenzie, J. E., Bossuyt, P. M., Boutron, I., Hoffmann, T. C., Mulrow, C. D., Shamseer, L., Tetzlaff, J. M., Akl, E. A., Brennan, S. E., et al. (2021) · Systematic Reviews · doi:10.1186/s13643-021-01626-4 · Cited by 13,874
Standards for Reporting Qualitative Research: A Synthesis of Recommendations
O’Brien, B. C., Harris, I. B., Beckman, T. J., Reed, D. A., & Cook, D. A. (2014) · Academic Medicine · doi:10.1097/acm.0000000000000388 · Cited by 12,211
CONSORT 2025 statement: updated guideline for reporting randomised trials
Hopewell, S., Chan, A.-W., Collins, G. S., Hróbjartsson, A., Moher, D., Schulz, K. F., Tunn, R., Aggarwal, R., Berkwits, M., Berlin, J. A., et al. (2025) · BMJ · doi:10.1136/bmj-2024-081123 · Cited by 1,195
The ethics of disclosing the use of artificial intelligence tools in writing scholarly manuscripts
Hosseini, M., Resnik, D. B., & Holmes, K. (2023) · Research Ethics · doi:10.1177/17470161231180449 · Cited by 341
Good Publication Practice (GPP) Guidelines for Company-Sponsored Biomedical Research: 2022 Update
DeTora, L. M., Toroser, D., Sykes, A., Vanderlinden, C., Plunkett, F. J., Lane, T., Hanekamp, E., Dormer, L., DiBiasi, F., Bridges, D., et al. (2022) · Annals of Internal Medicine · doi:10.7326/m22-1460 · Cited by 258
Ogrinc, G. S., Davies, L., Goodman, D., Batalden, P. B., Davidoff, F., & Stevens, D. (2015) · The Journal of Continuing Education in Nursing · doi:10.3928/00220124-20151020-02 · Cited by 195
Luo, X., Tham, Y. C., Giuffrè, M., Ranisch, R., Daher, M., Lam, K., Eriksen, A. V., Hsu, C.-W., Ozaki, A., Moraes, F. Y. D., et al. (2025) · BMJ Evidence-Based Medicine · doi:10.1136/bmjebm-2025-113825 · Cited by 97
Updating standards for reporting diagnostic accuracy: the development of STARD 2015
Korevaar, D. A., Cohen, J. F., Reitsma, J. B., Bruns, D. E., Gatsonis, C., Glasziou, P., Irwig, L., Moher, D., de Vet, H. C. W., Altman, D. G., et al. (2016) · Research Integrity and Peer Review · doi:10.1186/s41073-016-0014-7 · Cited by 88
Techniques for supercharging academic writing with generative AI
Lin, Z. (2024) · Nature Biomedical Engineering · doi:10.1038/s41551-024-01185-8 · Cited by 87
Research works printed but not linked: 1
- Matthew Uffindell. The History of the Tutorial. Greene’s College Oxford. https://greenesoxford.com/greenes-education/the-history-of-the-tutorial/.
Cited by
The permissibility of generative Artificial Intelligence in conference paper preparation
Spennemann, D. H. R. (2026) · Charles Sturt University Research Output (CRO) · doi:10.26189/1ab6f993-4fe4-4f9d-8008-f07f6e632f35 · Cited by 1
Mehta, N., Benjamin, J. C., Mehta, S., Valanci, S., MacNeill, H., & Masters, K. (2026) · Medical Teacher · doi:10.1080/0142159x.2026.2681971
Self-citationHesham, A., & Sharfeddin, A. (2026) · South Mediterranean University Journal of Pure and Applied Sciences · doi:10.67525/6pnc0868
Sharfeddin, A. (2026) · South Mediterranean University Journal of Pure and Applied Sciences · doi:10.67525/92f7hc64
Chandra, Y., & Tan, J. (2026) · Policy Studies Journal · doi:10.1111/psj.70165
Metadata
- Authors
- Mehta, N., Felder, R. M., Kodish, E., Imrey, P. B., & Frankel, R. M.
- Affiliations
- Cleveland Clinic Lerner College of Medicine (US)
- Cleveland Clinic (US)
- Indiana University School of Medicine (US)
- Corresponding author
- Mehta, Neil
- Venue
- Research Integrity and Peer Review
- Published
- 2026-05-10
- Volume
- 11
- Issue
- 1
- ISSN
- 2058-8615 (ISSN-L)
- PMC
- PMC13157658
- OpenAlex
- W7160719221
- Semantic Scholar
- 43aaf24b1a4bf6552e901074742b2fb9fc75189a
- Cited by
- 9
- References
- 15
- Citations by year
- 2026: 9
- FWCI
- 44.97
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- cc-by-nc-nd
- creativecommons.org/licenses/by-nc-nd/4.0 (tdm)
- creativecommons.org/licenses/by-nc-nd/4.0 (vor)
- Locations
- Research Integrity and Peer Review (publishedVersion, cc-by-nc-nd)
- PubMed Central (publishedVersion, cc-by-nc-nd)
- PubMed (publishedVersion)
- IUScholarWorks (Indiana University) (submittedVersion, cc-by-nc-nd)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Topics
- Social Sciences › Education › Academic integrity and plagiarism
- Social Sciences › Library and Information Sciences › Academic Writing and Publishing
- Social Sciences › Library and Information Sciences › Publishing and Scholarly Communication
- Keywords
- generative AI, large language models, AI disclosure, reporting guidelines, research integrity, scientific manuscript writing, AI-assisted research, AI-assisted writing, research ethics, publication ethics, conflict of interest disclosure, authorship, traceability, EQUATOR Network, journal editorial policy
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
A language model drew the claims from the paper on 2026-11-12, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-11-12 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.
