PETMALU-AI helps authors document how generative AI contributed to research writing and helps editors and reviewers evaluate whether that use has been reported transparently.
Authors document what occurred. Editors and reviewers evaluate whether applicable information is adequately reported.
Combine a materiality profile with one or more AI functions, manage multiple tools, document item-level reporting, and generate several manuscript-ready outputs.
Determine likely applicability, assess reporting, record evidence, and generate item-linked revision requests without a misleading pass/fail score.
The paper’s provisional levels establish reporting depth. The website adds a separate multi-select function layer—such as translation, literature synthesis, coding, synthetic data, or figures—to help users construct an initial applicable checklist.
Learn how applicability is determined →Minimal editorial, moderate writing, substantive process, data-generative, or no AI.
Choose all tasks that actually occurred.
Confirm, add, remove, document, or justify non-applicability.
The nine PETMALU-AI domains are grouped into three complementary dimensions that make human responsibility, process traceability, and the substantive use of AI visible.
Covers disclosure and scope, human contribution and responsibility, authorship compliance, ethical considerations, policy compliance, bias mitigation, and privacy safeguards.
Covers the AI system and version, access context and configuration, prompting and interaction records, output curation, reproducibility information, and supporting materials.
Covers AI-generated or transformed data and content, human verification, validation, error and bias review, robustness checks, limitations, reflexivity, and reproducibility constraints.
The official item title, description, rationale, and example originate from the validated framework. Plain-language explanations, additional examples, suggested reviewer comments, and scenario mappings are supplementary implementation aids.
Validated and supplementary item examples, scenario reports, and weak-to-comprehensive comparisons.
GuideWeb, DOCX, and PDF versions for authors, reviewers, journals, institutions, and educators.
TemplatesAI-use logs, prompt logs, output review, validation, privacy, checklist, and screening templates.
AdoptionPolicy language, submission fields, review questions, and screening workflow.
ReferenceDefinitions linked to items and guidance on where reporting may appear.
Static kitGenerate customizable instructions, submission questions, review guidance, FAQ, and adoption text.
Go beyond reading the framework by learning its concepts, practicing realistic scenarios, generating structured AI-use reports, and supporting editorial assessment—all directly in your browser with no account or installation required.
Learn each reporting domain through guided lessons, knowledge checks, progress tracking, and completion certificates.
12 casesPractice applying PETMALU-AI to realistic reporting scenarios with immediate feedback and model responses.
AuthorsCreate structured AI-use reports using customizable templates that can be adapted for manuscripts, supplementary materials, or institutional documentation.
EditorsGenerate structured reviewer comments, revision requests, editorial summaries, and verification reports based on the PETMALU-AI framework.
Please cite the published paper so readers can locate the framework’s development, validation, and official 33-item checklist.
Garcia, M. B. (2026). PETMALU-AI: A reporting checklist for transparent and accountable generative AI-assisted research writing. Education Sciences, 16(7), 1127. https://doi.org/10.3390/educsci16071127