Artificial Intelligence (AI) Usage Policy
Jurnal Phi: Jurnal Pendidikan Fisika dan Fisika Terapan is committed to maintaining the highest standards of academic integrity, originality, and ethical publishing practices. In response to the increasing integration of Artificial Intelligence (AI) tools in academic workflows, this policy outlines the acceptable and unacceptable uses of AI tools for authors, reviewers, and editors involved in the publication process.
This policy is formulated in line with international publication ethics guidelines, including recommendations from the Committee on Publication Ethics (COPE) and global standards for generative AI in scientific publishing. Specifically, Jurnal Phi adopts and adheres to the official policy standards established by Elsevier:
AI Policy Guidelines for Authors
Jurnal Phi: Jurnal Pendidikan Fisika dan Fisika Terapan outlines the following acceptable and unacceptable uses of AI tools for manuscript preparation.
A. Permitted Uses of AI
Authors may use AI tools with a maximum tolerance of 20% to assist with specific aspects of manuscript preparation:
- Language and Readability: AI tools may be employed for grammar correction, language refinement, and stylistic adjustments. Authors remain solely responsible for reviewing and verifying the accuracy of any changes.
- Illustrative Examples: Authors can include examples generated by AI for explanatory or illustrative purposes, provided these examples are clearly labeled and properly cited in the manuscript.
- Data Analysis: AI tools may support data analysis conducted through traditional physics or physics education research methods. However, the methodology must include a detailed description of the tools, parameters, and rationale for their use.
B. Prohibited Uses of AI
- Authorship and Content Creation: Authors must not use AI tools to generate substantive content for manuscripts, including sections such as abstracts, literature reviews, methodologies, results, or conclusions. Intellectual contributions must be entirely the original work of the human authors.
- Data Generation: AI tools cannot be used to fabricate or simulate raw data, statistical analyses, or experimental or survey findings. All reported data must originate from authentic and verifiable research.
- Image Creation: Images, figures, or visualizations generated by AI are not permitted due to concerns regarding authenticity, copyright, and ethical compliance.
C. Disclosure Requirement
Authors must explicitly disclose any use of AI tools in their manuscript. A disclosure statement should be included in the Acknowledgments section, specifying the exact tools used and their purpose.
Example Disclosure Statement:
"This manuscript was proofread using [AI Tool Name] for grammar and clarity improvement."
AI Policy Guidelines for Reviewers
Jurnal Phi: Jurnal Pendidikan Fisika dan Fisika Terapan outlines the following acceptable and unacceptable uses of AI tools for reviewers involved in the peer-review process.
A. Permitted Uses of AI
Reviewers may use AI tools solely to assist in the preparation of their review reports, such as:
- Language and Clarity Refinement: Refining grammar, sentence structure, or formatting of the review report to ensure constructive feedback is clearly communicated to authors.
- Reference and Fact Checking: Verifying references or basic factual scientific context relevant to the physics and physics education content of the manuscript.
B. Prohibited Uses of AI
- Autonomous Intellectual Evaluation: Reviewers must not use AI tools to evaluate the intellectual content, scientific novelty, methodology, or generate substantive portions of their review reports. The critical evaluation must remain entirely the reviewer's independent professional judgment.
- Breach of Confidentiality: Reviewers must strictly avoid uploading, pasting, or exposing any part of the manuscript to AI tools or online platforms, as this violates manuscript confidentiality and author copyright.
C. Reviewer Responsibilities
- Identifying AI Misuse: Reviewers are expected to identify and alert the editorial team regarding any suspected misuse of AI by authors, such as undeclared AI-generated text, fabricated experimental or survey data, or hallucinated citations.
- Upholding Ethical Standards: Reviewers must uphold strict confidentiality and ethical integrity throughout the entire peer-review process.
AI Policy Guidelines for Editors
Jurnal Phi: Jurnal Pendidikan Fisika dan Fisika Terapan outlines the following acceptable and unacceptable uses of AI tools for editors managing the publishing process.
A. Permitted Uses of AI
Editors may use AI tools for administrative and operational tasks, such as:
- Workflow Optimization: Streamlining editorial processes and tracking manuscript progress.
- Communication Enhancement: Enhancing language clarity in editorial decision letters or reviewer invitation emails.
- Substantive Evaluation with Human Oversight: Editors may utilize AI tools to assist in evaluating the intellectual merit of manuscripts or drafting correspondence, provided there is strict human oversight, review, and validation.
- Integrity Screening: Utilizing plagiarism and AI-detection tools to verify compliance with publication ethics.
B. Prohibited Uses of AI
- Autonomous Decision-Making: Editors must not delegate publishing decisions (accept, revise, reject) entirely to AI algorithms without independent, critical evaluation by the editorial board.
- Breach of Manuscript Confidentiality: Editors must not upload submitted manuscripts into public AI systems without explicit data protection protocols, safeguarding author intellectual property and confidentiality.
C. Editor Responsibilities
Editors are responsible for ensuring adherence to Jurnal Phi's AI policies by:
- Verifying Author Disclosures: Checking that authors explicitly disclose any AI usage and addressing potential ethical inconsistencies or violations.
- Preserving Peer-Review Rigor: Ensuring that the peer-review process remains rigorous, unbiased, and driven by expert critical thinking.
- Guiding Reviewers: Providing clear guidelines on AI-related ethical standards to reviewers and monitoring adherence across all review workflows.
Ethical Considerations for All Parties
AI tools can generate plausible-sounding but incorrect or fabricated information, a phenomenon commonly referred to as "AI hallucination." All parties involved in the publication process at Jurnal Phi: Jurnal Pendidikan Fisika dan Fisika Terapan—authors, reviewers, and editors—must independently verify and guarantee the accuracy, validity, and scientific integrity of any content influenced by AI.
Furthermore, the integration of AI tools must rigorously adhere to ethical publishing standards, including:
- Respect for Data Privacy: Ensuring that raw research data, personal identifiers, and confidential manuscripts are never exposed to public AI systems.
- Informed Consent: Safeguarding participant privacy and ethical compliance, particularly in physics education research involving human subjects, such as students and educators.
- Bias Mitigation: Actively identifying and mitigating algorithmic biases that AI models may introduce into scientific writing, analysis, or evaluation.
Declaration of Generative AI in Scientific Writing
Authors are required to explicitly disclose any utilization of generative AI technologies upon submitting a manuscript to Jurnal Phi: Jurnal Pendidikan Fisika dan Fisika Terapan. This policy specifically addresses the application of AI during the drafting and writing phases, distinct from AI tools employed for data analysis or insight extraction during the research phase.
- Permitted Scope: The use of Generative AI and AI-assisted tools is permitted solely for the purpose of enhancing the language quality and readability of the manuscript.
- Human Oversight: Authors must exercise rigorous human oversight when using these technologies. Because AI models can produce output that appears authoritative yet may be factually inaccurate, incomplete, or biased, authors are required to thoroughly review and edit any AI-generated content. Ultimately, the authors bear full responsibility and liability for the integrity and accuracy of the published work.
- Authorship Restrictions: Generative AI tools and AI-assisted technologies cannot be credited as an author or co-author. Authorship entails legal accountability, ethical responsibility, and intellectual contributions that are exclusive to human researchers.
- Placement Requirement: If generative AI has been used in the writing process, authors must include a disclosure statement immediately preceding the References section at the time of initial submission.
Declaration Format and Example
Heading: Declaration of Generative AI Usage in the Writing Process
"During the drafting of this manuscript, the author(s) utilized [NAME OF TOOL / SERVICE] for the purpose of [SPECIFIC REASON, e.g., refining sentence structure/translating text]. Following the use of this tool, the author(s) reviewed and revised the content as necessary and accept full responsibility for the final content of the article."
Note: This declaration requirement does not apply to standard tools used for basic spelling, routine grammar checking, or reference management (e.g., Mendeley or Zotero). If no generative AI was used, no statement is required. For complete guidelines, please refer to the Jurnal Phi Artificial Intelligence (AI) Usage Policy.
Peer-Review Confidentiality (Reviewers & Editors)
To safeguard research confidentiality and protect authors' intellectual property rights, reviewers and editors are strictly prohibited from uploading manuscripts into generative AI platforms, such as ChatGPT or similar services, during the peer-review and evaluation process. This aligns with our commitment to data privacy as outlined in our editorial policies.

