Generative AI Policies

Generative AI Policies

  

  1. Core Principles (The STM Integrity Hub)

Before using any AI tool (like ChatGPT, Claude, or Grammarly GO), understand the three pillars of academic integrity regarding AI:

  • Transparency: You must disclose the use of AI.
  • Accountability: Humans are fully responsible for the accuracy and integrity of the work, not the AI.
  • Originality: AI cannot hold copyright or authorship.
  1. Guidelines for Authors

If you are writing a manuscript for publication:

✅ You CAN use GenAI to:

  • Improve the readability and language of your work (grammar, flow).
  • Brainstorm ideas or structure the manuscript.
  • Summarize your own notes.

 

❌ You CANNOT use GenAI to:

  • Replace Authorship: AI cannot be listed as a co-author. It cannot take legal responsibility or sign an author agreement.
  • Interpret Data: Do not rely on AI to draw scientific conclusions or interpret data, as it is prone to "hallucinations" (generating false facts).
  • Generate Synthetic Data: Unless the paper is specifically about AI methodology, do not use AI to fabricate data.
  • Create Images: Elsevier generally does not permit the use of GenAI to create or alter images/figures/charts, except for minor brightness/contrast adjustments or unless specifically allowed by the target journal (check specific instructions).
  •  

⚠️ The Golden Rule: Disclosure

If you use AI, you must add a statement at the end of your manuscript (usually before References).

Sample Declaration:

"During the preparation of this work, the author(s) used [Name of Tool / Service] in order to [Reason: e.g., improve readability and language]. After using this tool/service, the author(s) reviewed and edited the content as needed and take(s) full responsibility for the content of the publication."

  1. Guidelines for Reviewers

If you are reviewing a paper for a journal or a student's thesis:

❌ STRICT PROHIBITION:

  • Do not upload the manuscript to a GenAI tool.
    • Uploading a manuscript to a public AI (like ChatGPT) is a violation of confidentiality. These tools may use the data to train their models, effectively leaking the author's unpublished work.

✅ You CAN use GenAI to:

  • Polish the language of your own review report (after you have written the critique yourself), provided you do not input the specific details of the manuscript into the prompt.
  1. Guidelines for Editors

For senior lecturers and professors managing journals:

  • Human Decision Making: An AI tool cannot make the final decision to Accept or Reject a manuscript. Editorial decisions must be human-led.
  • Screening Tools: Editors are permitted to use AI-based tools for screening (e.g., checking for plagiarism or image manipulation), but the results must be verified by a human.
  • Correspondence: AI can assist in drafting communication letters, but the editor is responsible for the tone and accuracy.

Summary Table: Roles & Responsibilities

Feature

Authors

Reviewers

Editors

Can be an Author?

No

N/A

N/A

Can use for Polish/Grammar?

Yes (Must Disclose)

Yes (Caution on privacy)

Yes

Upload Manuscript to AI?

N/A

Strictly Forbidden

Internal tools only

Accountability

100% Human

100% Human

100% Human


These guidelines based on Elsevier & STM Association Guidelines