Ethical Guidelines for Using AI in Academic Writing

Creative Science recognizes that artificial intelligence (AI), including generative AI and AI-assisted tools, can support research and scholarly writing. However, its use must not compromise scientific integrity, accuracy, originality, confidentiality, or human accountability.

These guidelines apply to authors, reviewers, editors, and editorial staff involved in the journal's publication process.

Creative Science recognizes that artificial intelligence (AI), including generative AI and AI-assisted tools, can support research and scholarly writing. However, its use must not compromise scientific integrity, accuracy, originality, confidentiality, or human accountability.

These guidelines apply to authors, reviewers, editors, and editorial staff involved in the journal's publication process.

General Principles

The responsible use of AI must follow six principles:

  1. Human accountability: Humans remain responsible for all research content, scholarly judgments, and editorial decisions.
  2. Confidentiality and privacy: Protected information must not be disclosed to unauthorized AI systems.
  3. Transparency: Material AI assistance must be clearly disclosed.
  4. Accuracy: AI-generated outputs must be independently reviewed and verified.
  5. Research integrity: AI must not be used to fabricate, falsify, plagiarize, or misrepresent scholarly work.
  6. Responsible use: AI tools must be used in accordance with applicable ethical, legal, institutional, and journal requirements.

Permitted Uses of AI by Authors

Authors may use AI-assisted tools to support legitimate research and writing activities, including:

  1. Improving grammar, spelling, readability, and language clarity.
  2. Assisting with translation and language editing.
  3. Organizing manuscript structure or preparing preliminary drafts.
  4. Exploring relevant literature, provided that all references and source information are independently verified.
  5. Supporting coding, data processing, statistical analysis, or other research tasks when the methods and results are appropriately validated.
  6. Preparing explanatory illustrations or other non-evidentiary visual materials, subject to appropriate disclosure and copyright requirements.

AI assistance does not replace the author's responsibility to evaluate the accuracy, originality, and scientific validity of the final manuscript.

Prohibited Uses of AI

Authors must not use AI tools to:

  1. Fabricate or falsify research data, observations, methods, results, or conclusions.
  2. Generate fictitious references, DOI identifiers, quotations, or other source information and present them as authentic.
  3. Create or manipulate scientific images in ways that misrepresent experimental or observational evidence.
  4. Produce plagiarized content or conceal the use of unattributed third-party material.
  5. Misrepresent unverified AI-generated analysis as validated research findings.
  6. Upload confidential, sensitive, or personally identifiable research information to unauthorized AI systems.
  7. Circumvent ethics approval, informed consent, or other research-integrity requirements.
  8. Conceal AI assistance when disclosure is required under this policy.

AI Authorship and Accountability

AI tools cannot be listed as authors or co-authors because they cannot take responsibility for the accuracy, originality, integrity, or ethical implications of scholarly work.

All listed human authors remain fully responsible for the manuscript, including any AI-assisted content.

Authors must verify factual claims, references, calculations, analyses, interpretations, and visual materials before submission. The corresponding author is responsible for ensuring that the AI declaration is complete and accurate on behalf of all authors.

Disclosure of AI Use

Authors must disclose the use of generative AI or AI-assisted tools when these tools materially contribute to manuscript preparation, content generation, research methods, data analysis, interpretation, or other scholarly outputs.

The disclosure should include, where applicable:

  1. The name of the AI tool or model and its version, if known.
  2. The purpose of using the tool.
  3. The stage of research or manuscript preparation in which it was used.
  4. The nature and extent of AI assistance.
  5. The steps taken by the authors to review, verify, or validate the outputs.

The journal may request additional information when necessary to assess the transparency or reproducibility of the work.

Placement of AI Disclosures

A. AI Used in Manuscript Preparation

When AI is used to assist with writing, editing, translation, or manuscript organization, authors should include a declaration under the heading:

Declaration of Generative AI and AI-Assisted Technologies in the Writing Process

The declaration should be placed in the manuscript's declarations section or another location specified in the journal's author guidelines.

B. AI Used in Research Methodology

When AI is used to collect, generate, process, classify, analyze, or interpret research data, authors must describe its methodological role in the Materials and Methods section or an equivalent section.

The description should include relevant details of the tool or model, version, developer, input data, procedures, validation methods, and limitations.

Information should be sufficient to allow readers and reviewers to evaluate the reliability and reproducibility of the research.

C. AI-Generated or AI-Assisted Figures

AI-generated or materially AI-modified illustrations must be clearly identified in the figure caption or an accompanying statement.

AI-generated illustrations must not be presented as authentic experimental images or observational evidence.

Any use of AI in scientific image processing must preserve the integrity of the underlying data and be described when necessary.

Examples of AI Disclosure Statements

Authors should adapt the following examples to reflect their actual use of AI.

Example 1: Language Editing

"During the preparation of this manuscript, the authors used [AI Tool/Model] to assist with grammar correction and language improvement. All AI-assisted text was reviewed and revised by the authors, who take full responsibility for the final content."

Example 2: Translation and Manuscript Organization

"The authors used [AI Tool/Model] to assist with translation and the organization of selected manuscript sections. The authors reviewed the outputs, verified their accuracy, and approved the final manuscript."

Example 3: Substantive Writing Assistance

"During manuscript preparation, the authors used [AI Tool/Model] to assist with drafting and restructuring portions of the [specified sections]. All generated content, factual claims, and references were independently reviewed, verified, and revised by the authors. The authors accept full responsibility for the final manuscript."

Example 4: AI Used in Research Methodology

"An AI-based system, [Tool/Model, Version, Developer], was used for [specific research purpose]. The system processed [description of inputs] using [summary of procedures]. The outputs were evaluated using [validation method]. Further details and limitations are provided in the Materials and Methods section."

Example 5: AI-Assisted Illustration

"Figure [Number] was created with the assistance of [AI Tool/Model] for illustrative purposes. The image does not represent original experimental or observational data."

Example 6: No Generative AI Use

"The authors declare that no generative AI or AI-assisted technologies were used in the preparation of this manuscript."

Authors must provide an accurate declaration. A statement of no AI use must not be submitted when relevant AI assistance occurred.

AI Use by Peer Reviewers

Reviewers must maintain the confidentiality of all manuscripts and related materials.

Reviewers shall:

  1. Not upload unpublished manuscripts, figures, tables, supplementary files, or confidential editorial correspondence to publicly accessible or unauthorized AI systems.
  2. Not delegate substantive manuscript evaluation or peer-review judgments to AI tools.
  3. Use AI assistance for limited language refinement or organizational support only when manuscript confidentiality is protected and the use complies with journal policy.
  4. Independently verify any AI-assisted content included in their review reports.
  5. Remain fully responsible for the accuracy, fairness, originality, and confidentiality of their reviews.
  6. Disclose relevant AI assistance when requested by the editorial office.

AI Use by Editors and Editorial Staff

Creative Science permits the responsible use of generative AI to support selected editorial and publication tasks, including:

  1. Editorial Correspondence: Assisting with the drafting and language refinement of routine emails and administrative correspondence.
  2. Cover Artwork: Creating illustrative cover artwork based on graphical abstracts or research summaries, while preserving the scientific meaning of the original work.

The journal uses Turnitin for manuscript similarity screening. Similarity reports are reviewed by the editorial team and are not treated as conclusive evidence of plagiarism.

All AI-generated content must be reviewed and verified by the editorial team before use. AI-generated cover artwork must not misrepresent research findings or infringe copyright or other intellectual property rights.

Generative AI must not replace human judgment in manuscript evaluation, peer review, publication decisions, or investigations of research misconduct.

Unpublished manuscripts, confidential peer-review reports, personal information, and other protected materials must not be uploaded to unauthorized AI systems.

Editors and editorial staff remain responsible for all AI-assisted editorial and publication activities.

Noncompliance with the AI Policy

Suspected violations of this policy will be assessed according to the nature and seriousness of the concern, the available evidence, and its impact on research or publication integrity.

The journal may request clarification, additional disclosure, corrections, or supporting documentation.

Serious violations, including fabricated references, falsified data, misleading scientific images, or unauthorized disclosure of confidential information, may result in manuscript rejection or post-publication action.

Failure to disclose AI use will be evaluated on a case-by-case basis and does not automatically result in retraction.

Editorial actions will follow the journal's procedures for dealing with research misconduct.

Policy Review

Creative Science may review and update these guidelines as AI technologies, scholarly publishing practices, and relevant ethical standards evolve.

Authors, reviewers, and editors are responsible for following the version of the policy in effect during their participation in the publication process. Each revision will be issued with an updated version number and effective date.