Artificial Intelligence Policy

Artificial Intelligence (AI) Policy

Classroom Experiences (CARE)

This policy governs the use of Artificial Intelligence (AI) and AI-assisted tools in manuscripts submitted to Classroom Experiences (CARE). Our goals are transparency, accountability, research integrity, and the protection of privacy and intellectual property. In line with leading publishers and editorial bodies, AI tools cannot be authors, and any use of AI must be clearly disclosed and verified by humans.

1) Scope & Principles

  • Transparency: Authors must disclose what tool was used, which version, when, for what purpose, and which sections were affected.
  • Human accountability: Human authors remain fully responsible for the accuracy, originality, ethics, and legality of all content.
  • Privacy & IP protection: Confidential or proprietary materials and personal data must not be uploaded to public AI tools.

2) Definitions

AI / AI-assisted tools include systems that can generate, edit, summarize, or refactor text, data, images, or code (e.g., large language models or image generators). Disclosure is the author’s written statement (see templates below) specifying the tools, versions, purposes, affected sections, and the human verification applied.

3) Authorship & Responsibility

  1. AI cannot be an author or corresponding author. AI tools lack authorship criteria, legal standing, and accountability.
  2. Human authors must ensure no plagiarism or fabricated content (including fabricated citations or images) is introduced via AI outputs.
  3. Authors must be able to attest that all cited sources exist and have been appropriately interpreted and referenced.

4) Use of AI by Authors: Allowed vs. Prohibited

4.1 Allowed with Disclosure & Human Verification

  • Language editing (grammar, clarity), limited paraphrasing, or formatting support.
  • Early ideation (e.g., outlining) that is subsequently rewritten, fact-checked, and verified by the authors.
  • Consistency checks (references, style) with manual confirmation by authors.

4.2 Prohibited

  • Listing AI as an author or citing AI as an author.
  • Using AI to generate, alter, or fabricate original research data (e.g., experiments, classroom observations, surveys), unless explicitly permitted by the editor with complete methodological transparency and ethics approvals.
  • Uploading confidential, proprietary, or identifiable data to public AI systems.
  • Submitting AI-generated figures/images unless explicitly approved by the editor and clearly labeled; legal ownership and provenance must be documented.

5) Where to Disclose & What to Keep

  • Place the “AI Disclosure” statement in Acknowledgements or Methods (when AI influenced analysis or writing).
  • Maintain internal records (e.g., key prompts and outputs used) for possible editorial audit. Do not include raw prompts/outputs in the submission unless requested.

6) Data, Code, and Images

If AI supports analysis (e.g., code refactoring, topic modeling), authors must document pipelines, key parameters, and human validation. AI-generated images/figures are generally disallowed unless the editor grants a written exception; any permitted AI-generated visuals must be clearly labeled in captions with provenance details.

7) AI Use in Peer Review (Reviewers)

  • Manuscripts are confidential. Reviewers must not upload any manuscript content to public AI tools.
  • If a reviewer drafts comments with AI assistance, they remain responsible for accuracy, tone, and confidentiality; no fabricated citations or disclosures of private content.

8) Editorial Practice & Confidentiality

  • Editors must not upload confidential submissions to public AI services.
  • CARE may use licensed in-house tools (e.g., similarity checking, integrity screening) that preserve confidentiality and comply with applicable laws.

9) Detection, Audit, and Non-Compliance

AI-text detectors are imperfect; editorial decisions rely on substantive evidence (e.g., mismatched data, fabricated references, undisclosed AI). Potential actions include requests for clarification and logs, editorial notes, corrections, or retractions, aligned with industry ethics guidance.

10) Rights, Licensing, and Privacy

  • Verify the AI tool’s terms of use regarding data retention, training, and IP. Authors are responsible for ensuring that confidential or personal data are handled lawfully and ethically.
  • Authors retain responsibility for permissions and copyright of all materials, including any AI-assisted content.

11) Alignment with International Standards

CARE aligns with guidance from major editorial bodies and publishers regarding AI and authorship, disclosure, peer review confidentiality, and restrictions on AI-generated images.

12) Policy Updates

This policy is reviewed periodically to reflect evolving law, technology, and community practice. Revisions take effect upon publication on the CARE website.

Appendix A — AI Disclosure Templates

1) Language Editing Only

The authors used [Tool, version, access date] solely for grammar and language clarity. All scientific content, data analysis, and conclusions were authored and verified by the authors. No AI tools were used to generate or alter research data, results, or images.

2) Outline / Drafting Assistance (Rewritten by Authors)

The authors consulted [Tool, version, access date] to assist with initial outlining and phrasing for Sections [specify]. All passages were subsequently rewritten, fact-checked, and verified by the authors. No AI was used to generate or alter research data, results, or images.

3) Analytical Support (Non-data generation)

[Tool, version, access date] supported code refactoring/documentation for [method]. Parameter choices, validation, and interpretation were performed by the authors. Data and results were not generated by AI.

Appendix B — Quick Author Checklist

  • [  ] AI is not listed as an author and is not cited as an author.
  • [  ] AI use is disclosed with tool, version, date, purpose, affected sections, and human verification.
  • [  ] No data/images are generated or manipulated by AI without explicit editorial approval and clear labeling.
  • [  ] I keep internal records (e.g., key prompts/outputs) for possible editorial audit.
  • [  ] I checked the AI tool’s terms regarding data retention, training, and IP; no confidential or personal data were exposed.