CPRT welcomes responsible engagement with emerging technologies while maintaining the principle that human authors are accountable for the originality, accuracy, methodological soundness, intellectual contribution, and integrity of their submitted research. AI-assisted content must be subject to appropriate human evaluation, independent verification, and scholarly revision.

This policy establishes the journal's requirements concerning AI authorship, permissible assistance, mandatory disclosure, prohibited practices, accuracy and verification, humanities-specific research integrity, editorial oversight, and responsible innovation. It applies to authors submitting manuscripts to CPRT and, where relevant, to the handling of unpublished materials by editors and reviewers.

Authorship and Accountability

AI tools, generative AI systems, and large language models cannot be listed as authors or co-authors of manuscripts submitted to CPRT. Authorship represents substantive intellectual responsibility and requires the capacity to approve a manuscript, respond to scholarly criticism, disclose relevant interests, and accept accountability for the published work.

The responsibilities associated with scholarly authorship include:

• Intellectual responsibility: Developing, critically evaluating, and defending the research questions, arguments, interpretations, and conclusions presented in the manuscript.
• Accountability for content: Ensuring the accuracy, originality, reliability, and appropriate attribution of all submitted material.
• Scholarly responsiveness: Responding substantively to editorial enquiries, peer-review comments, and legitimate criticism concerning the research.
• Ethical compliance: Observing applicable requirements concerning research integrity, copyright, confidentiality, authorship, and responsible scholarly conduct.
• Approval and responsibility: Reviewing and approving the submitted and final published versions and accepting appropriate responsibility for the work.

These obligations cannot be delegated to AI systems. All listed authors remain responsible for the submitted manuscript, including its AI-assisted elements, and must be able to explain and justify the scholarly content for which they accept responsibility.

The use of AI does not establish authorship eligibility for an individual who has not made a qualifying intellectual contribution. Author lists must accurately reflect the participation and responsibilities of the human contributors, in accordance with the journal's Publication Ethics Policy.

Permitted Uses of AI

CPRT permits the responsible use of AI-assisted technologies for limited supporting activities in scholarly research and manuscript preparation. Permissible assistance must remain subordinate to the authors' independent intellectual judgement and must not compromise originality, transparency, source accuracy, or research integrity.

Examples of permitted supporting activities include:

• Language refinement: Improving the clarity, readability, grammatical accuracy, and linguistic consistency of author-prepared text.
• Grammar and proofreading assistance: Identifying possible typographical, syntactic, and mechanical errors for subsequent author review.
• Structural suggestions: Proposing possible organisational improvements to an existing manuscript without determining the substantive content of the scholarly argument.
• Preliminary outlines: Assisting with early-stage planning and the organisation of research ideas, subject to independent scholarly development by the authors.
• Summarisation support: Preparing preliminary summaries of publicly available information that authors subsequently verify against the original sources.

AI assistance must not substitute for the following core scholarly responsibilities:

• Original intellectual contribution and research design.
• Conceptual and philosophical analysis.
• Comparative argumentation and theoretical interpretation.
• Critical examination of primary and secondary sources.
• Independent scholarly judgement and responsibility for conclusions.

AI-generated suggestions concerning philosophical arguments, theoretical frameworks, or original research claims must not be accepted uncritically. Authors must independently assess, substantiate, revise, or reject such suggestions and ensure that their final manuscript represents a defensible human-authored scholarly contribution.

AI as a Research Tool or Object of Study

Where AI constitutes an explicit research instrument, object of philosophical investigation, or component of an interdisciplinary methodology, authors must explain its role with sufficient clarity to permit scholarly assessment. The relevant methods should identify the tools used, their functions, the nature of any generated materials, and the procedures adopted to evaluate the resulting evidence.

Research that critically examines AI-generated philosophical or religious discourse may analyse such outputs as research materials. However, AI-generated statements must be distinguished from independently verified historical facts, authentic textual sources, and established scholarly interpretations.

Mandatory Disclosure

Authors who use AI-assisted tools in preparing a manuscript must disclose that use in a dedicated statement at the time of submission. The disclosure must accurately describe the involvement of AI technologies and should be sufficiently specific to allow editors and readers to understand their role in the preparation of the scholarly work.

An AI-use disclosure should identify:

• Tool identification: The name of the AI tool or system used and its relevant version, where available.
• Purpose of use: The specific activities for which AI assistance was employed.
• Extent of involvement: The relevant stages or components of manuscript preparation, research processing, translation, or analysis in which AI was used.
• Human verification: The steps undertaken by the authors to assess, correct, and independently verify AI-assisted material.
• Scholarly responsibility: Confirmation that the authors retain responsibility for the final content, arguments, references, and conclusions.

Example AI-Use Disclosure

The following statement illustrates an appropriate disclosure where AI has been used for language refinement and editorial assistance. Authors should adapt the wording to describe their actual use.

Where AI has been used for substantive research procedures, including text classification, corpus processing, translation analysis, or the generation of research materials, authors should additionally explain the methodological role of the relevant tools in the manuscript itself. A general acknowledgement of AI assistance is not a substitute for an adequate description of research methods.

Authors should provide the disclosure separately from the anonymised manuscript when its inclusion could compromise double-anonymous peer review. The final disclosure may be incorporated into the publication record at the appropriate editorial or production stage.

Failure to disclose material AI involvement, or providing misleading information about its use, may constitute a breach of the journal's publication-ethics requirements. Where concerns arise, the editorial office may request clarification, supporting information, or a corrected declaration.

Prohibited Practices

CPRT prohibits practices in which AI technologies are used to misrepresent authorship, conceal the provenance of scholarly content, fabricate research evidence, compromise the integrity of peer review, or evade established requirements for responsible academic conduct.

The following practices are prohibited:

• AI authorship: Listing an AI tool or system as an author or co-author of a submitted manuscript.
• Unaccountable AI-generated manuscripts: Submitting substantially AI-generated scholarly content without adequate independent human evaluation, verification, intellectual development, and disclosure.
• Fabricated references: Including nonexistent publications, invented quotations, false citations, or inaccurately attributed arguments generated or suggested by AI.
• Fabrication or falsification: Using AI to create or alter research data, textual evidence, images, documentary materials, or findings in a manner that misrepresents authentic evidence.
• AI-assisted plagiarism: Using automated rewriting, paraphrasing, or translation to conceal the unacknowledged use of another person's scholarship.
• Peer-review manipulation: Using AI to create deceptive reviewer identities, prepare misleading reviewer communications, manipulate assessments, or otherwise interfere improperly with editorial decision-making.
• Unauthorised disclosure: Submitting confidential manuscripts, unpublished research materials, identifiable participant information, or protected editorial communications to external AI services without appropriate authorisation and safeguards.
• Misleading disclosure: Concealing material AI involvement or falsely describing the extent of human verification and intellectual contribution.

The use of AI to fabricate a historical source, construct a fictitious scholarly citation, or attribute an invented philosophical argument to an actual thinker is incompatible with the journal's standards of research integrity. Such conduct is particularly serious where the fabricated material forms a central part of the manuscript's evidence or conclusions.

Suspected violations may result in requests for clarification, suspension of editorial processing, manuscript rejection, or further ethical investigation. Where a serious violation is established after publication, the journal may issue a correction, expression of concern, or retraction, as appropriate under its Publication Ethics Policy.

Accuracy and Verification

Authors bear responsibility for independently verifying the reliability of all AI-assisted content included in their manuscripts. AI outputs may contain factual errors, fabricated references, inaccurate quotations, misleading summaries, or interpretations that lack sufficient textual or scholarly support.

Authors must verify, as relevant:

• Bibliographic accuracy: The existence, authorship, title, publication details, and identifying information of cited works.
• Quotation authenticity: The exact wording, source location, and proper attribution of passages quoted in the manuscript.
• Translation accuracy: The relationship between original-language materials and translations or transliterations used in the analysis.
• Historical and factual claims: The evidential basis of relevant claims concerning intellectual traditions, historical contexts, philosophical schools, and religious practices.
• Conceptual coherence: The logical consistency, originality, and appropriate scholarly justification of arguments influenced by AI-generated suggestions.
• Research evidence: The reliability of any AI-assisted extraction, classification, coding, summarisation, or analysis of research materials.

References suggested by AI tools must be checked against the publications or source records actually consulted. Authors should not treat an AI-generated bibliography, quotation, or source summary as verified evidence merely because it appears plausible or includes detailed bibliographic information.

Where AI-assisted translation or summarisation affects a central philosophical claim, authors should compare the output with the relevant original-language text, authoritative edition, or established scholarly translation as appropriate to the research. Significant ambiguities, contested terminology, and alternative readings should be acknowledged where relevant.

The final manuscript must reflect the authors' independent evaluation of the evidence and their responsibility for the conclusions presented. AI assistance does not reduce the journal's requirements concerning source accuracy, citation integrity, or the originality of scholarly argumentation.

Research Integrity in Humanities Scholarship

Comparative philosophy and religious studies require sustained attention to conceptual distinctions, textual transmission, historical interpretation, linguistic context, and the intellectual diversity of philosophical and religious traditions. AI tools may produce fluent summaries or comparisons without adequately reflecting the complexity of the underlying sources.

Authors using AI technologies in humanities scholarship must maintain:

• Careful textual interpretation: Arguments concerning philosophical and religious texts must be grounded in accurate readings of the relevant primary materials.
• Accurate representation of traditions: Authors must avoid unsupported generalisations, artificial equivalences, or misleading descriptions of philosophical schools and religious communities.
• Cultural and historical sensitivity: Scholarship should account for differences in language, historical circumstances, cultural setting, and modes of textual interpretation.
• Transparent methodological reasoning: Comparative and interpretive choices must be justified through scholarly argument rather than accepted solely on the basis of automated outputs.
• Source and translation accountability: Authors must accurately identify the editions, translations, commentaries, and textual witnesses upon which their analysis depends.
• Recognition of interpretive plurality: Where relevant, manuscripts should acknowledge contested readings, competing scholarly positions, and limitations in comparative interpretation.

Classical Texts and Translation

AI-generated translations of classical philosophical works, religious scriptures, commentaries, or historically significant terminology require particular scrutiny. Such outputs may overlook established textual conventions, obscure grammatical ambiguities, or present contested philosophical concepts as straightforward equivalents.

Authors should identify the textual sources and translations relied upon, explain consequential translation choices, and distinguish established scholarly interpretations from their own analytical claims. Where an AI-assisted translation materially informs the research, the use of the tool must be disclosed and the relevant text independently evaluated.

Comparative Interpretation and Conceptual Equivalence

Comparative arguments often involve concepts whose meanings are shaped by distinct linguistic, historical, philosophical, and religious contexts. AI-assisted comparisons must be examined critically to determine whether apparent similarities rest upon defensible conceptual relationships rather than superficial terminological resemblance.

The authors are responsible for establishing the basis of comparison, assessing relevant differences, and explaining the significance of their conclusions. Automated suggestions do not replace the interpretive work required to produce responsible and original comparative scholarship.

Editorial Oversight

The editorial team is responsible for assessing concerns relating to the inappropriate use or disclosure of AI technologies in submitted and published scholarly work. Such concerns are considered in accordance with the journal's Publication Ethics Policy and its responsibilities for fair, evidence-based editorial assessment.

Where appropriate, the editorial team may:

• Request clarification concerning the tools used, their functions, and the extent of AI involvement.
• Request corrected or more detailed AI-use declarations where existing disclosures are incomplete or misleading.
• Ask authors to verify specific quotations, references, translations, factual claims, or other AI-assisted content.
• Seek relevant documentation or explanations concerning AI-assisted research methods where necessary for scholarly assessment.
• Investigate substantiated concerns involving fabricated sources, plagiarism, research misrepresentation, confidentiality breaches, or manipulation of peer review.
• Apply proportionate editorial or post-publication measures where an ethical violation has been established.

Editorial assessment of suspected AI misuse should consider the available evidence, the nature of the concerns, the authors' explanations, and the potential effect on the reliability of the manuscript. Automated AI-content detection results, if considered, must not be treated as conclusive evidence of misconduct without appropriate independent evaluation.

Confidentiality in Editorial and Peer Review

Editors and reviewers must preserve the confidentiality of unpublished manuscripts, review reports, editorial correspondence, and other materials received through the journal's editorial process. Such materials must not be uploaded or disclosed to external AI services without appropriate authorisation and safeguards protecting confidentiality, intellectual property, and applicable data rights.

Reviewers remain personally responsible for the scholarly content of their assessments. AI-generated review comments must not be substituted for an independent expert evaluation, and AI technologies must not be used in ways that compromise the integrity or confidentiality of the review process.

Where an ethical concern is identified, the journal may seek clarification from the relevant parties and determine an appropriate response through its established editorial procedures. Any corrective action should reflect the seriousness of the concern and its effect on the integrity of the scholarly record.

Commitment to Responsible Innovation

CPRT supports responsible and transparent engagement with emerging technologies in academic research. The journal recognises that AI developments raise important philosophical questions concerning knowledge, authorship, interpretation, reasoning, ethics, cultural representation, and the relationship between human intellectual agency and computational systems.

The journal encourages scholars to investigate these questions critically while maintaining established standards of source verification, methodological transparency, conceptual analysis, and academic responsibility. Responsible innovation requires the acknowledgement of technological capabilities and limitations, together with continued attention to the intellectual and ethical obligations of scholarly authorship.

CPRT may review and update this policy as scholarly practices, technological capabilities, and relevant ethical standards evolve. Any revision should preserve the journal's underlying commitments to intellectual originality, responsible research conduct, editorial independence, and the integrity of published scholarship.