AI Policy

Generative AI Policy

Generative Artificial Intelligence (AI) tools, including large language models (LLMs) and multimodal models, are rapidly evolving and finding applications across business, consumer, and academic sectors.

The Cognitive Insight in Education acknowledges the potential opportunities presented by Generative AI tools. These include facilitating initial idea generation and exploration, assisting authors in articulating content in a non-native language, and accelerating the research and dissemination process.

However, the Journal also recognizes the significant risks inherent in current Generative AI technologies. This policy guides authors, editors, and reviewers to ensure the responsible and ethical use of these tools, upholding the core values of academic integrity, accuracy, and originality essential to social and scientific education.

Primary Risks of Generative AI Tools

Generative AI tools (such as ChatGPT, Copilot, Gemini, Claude, DALL-E, Midjourney, etc.) can produce diverse content forms, including text, images, audio, and synthetic data. Their use in a scholarly context must be governed by stringent ethical standards due to several primary risks:

  • Inaccuracy and Bias: As statistically driven systems, Generative AI tools can introduce inaccuracies, falsities ("hallucinations"), or biases that are difficult to detect, verify, and correct, posing a particular threat to scientific and educational rigor.
  • Lack of Attribution: These tools often fail to adhere to the scholarly convention of accurately attributing ideas, quotes, and citations, potentially leading to plagiarism and misrepresentation of intellectual property.
  • Confidentiality and Intellectual Property Risks: Presently, many Generative AI tools operate on third-party platforms that may not provide adequate standards of confidentiality, data security, or copyright protection for unpublished scholarly work.
  • Unauthorized Data Use: AI providers may repurpose user-inputted or outputted data for model training, potentially infringing upon the rights of authors and publishers.

 

Guidelines for Authors

Authors are ultimately accountable for the originality, validity, and integrity of their submissions. The use of Generative AI tools is expected to be responsible, transparent, and in strict adherence to the Journal's policies on authorship and publishing ethics.

Permissible Uses (with Caution and Oversight):

  • Idea generation and initial exploration.
  • Language polishing and editing for clarity (particularly for non-native speakers).
  • Preliminary literature classification.
  • Coding assistance for data analysis.

Authors are required to critically review, verify, and revise any AI-generated output thoroughly. Scientific accuracy and methodological soundness remain a fundamental human responsibility. Generative AI tools must not be listed as an author.

Transparency Disclosure Requirement:

Authors must explicitly acknowledge any use of Generative AI tools within their manuscript via a statement in the Methods or Acknowledgments section, including:

  • The full name and version number of the tool used (e.g., ChatGPT-4).
  • A description of how and the purpose for which the tool was used.

Prohibited Uses:

  • Text or Code Generation without Rigorous Human Revision: AI must not be used to draft substantial portions of the manuscript without significant intellectual input.
  • Synthetic Data Generation: AI must not be used to generate or substitute missing research data.
  • Creation of Images, Figures, and Research Data: The journal prohibits the use of Generative AI to create or manipulate images, figures, charts, data tables, formulas, or original research data.
  • Generation of Abstracts or Supplemental Materials: AI must not be used to produce abstracts or supplementary materials.

 

Guidelines for Editors & Reviewers

The Cognitive Insight in Education is committed to the highest standards of editorial integrity and confidentiality.

Editors and Peer Reviewers are strictly prohibited from uploading any unpublished manuscripts, associated files, images, or information into Generative AI tools. Such actions breach confidentiality, violate intellectual property rights, and compromise the peer review process.

Peer Reviewers are selected for their subject expertise and are expected to provide original, critical assessments. The use of AI to analyze, summarize, or draft reviews of submitted manuscripts is not permitted. The sole exception is for language polishing of the review itself, on the condition that the reviewer remains entirely responsible for the accuracy, content, and integrity of their assessment.