Generative Al policies

Declaration of generative Artificial Intelligence (AI) and automated tools

  • For authors:
    • Authors should disclose the use of generative AI in preparing the article, beyond straightforward language correction, editing and formatting.
    • Authors are responsible for checking the validity of the output of any automated tools used in their research and preparing their manuscript.
    • Automated tools cannot be credited as authors.
    • Generative AI cannot be cited as a source.
    • The use of AI tools in the manuscript preparation process must be declared by adding a statement at the end of the manuscript when the paper is first submitted. The statement will appear in the published work and should be placed in a new section before the reference list. An example:

Title of new section: Declaration of generative AI and AI-assisted technologies in the manuscript preparation process.

Statement: During the preparation of this work the author(s) used [NAME OF TOOL/SERVICE] to [REASON]. 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 published article.

 

  • The declaration does not apply to the use of basic tools, such as tools used to check grammar, spelling, and references. In addition, authors are not required to disclose the use of AI tools or features within specialist disability-related assistive technology, provided these are used solely for accessibility purposes.
  • If you have nothing to disclose, you do not need to add a statement.
  • For journals:
    • Peer reviewers and editors should not use generative AI to create their assessments, due to risks such as breaches of confidentiality, superficial and non-specific feedback, bias, hidden prompts, and false information such as fake references; editing and rewriting may be acceptable if disclosed.
    • Any routine use of automated tools by the journal or publisher should be disclosed. The tool should have been appropriately tested.
    • Use of automated tools should be overseen by humans (human in the loop). Journals should ensure that an editor or other staff will verify the automated detection of integrity issues such as text similarity, figure manipulation or duplication, or undeclared generative AI use, or automated peer reviewer suggestions.
    • Disclosure: Reviewers should disclose AI use in their review reports, including the tool used and purpose of the use. Basic checks for spelling and grammar do not need to be disclosed.

Suggested disclosure statement: During the preparation of this report, I used [NAME OF TOOL / SERVICE] in order to [REASON]. After using this tool/service, I reviewed and edited the content as needed and I take full responsibility for its content.

 

What is a private AI tool?

When we refer to a "private" AI tool, we mean a version of an AI tool that does not retain, re-use, share, or learn from the content you submit to it. This means that when you paste or upload any content into the tool, that content should not be retained beyond what is necessary to provide the service, and must not be used to train or improve the AI model or shared with any third party.

Before using any AI tool, you should verify the following:

  • Read the privacy policy and terms of service. Look specifically for statements about data retention and model training. A private tool will clearly state that user inputs are not retained beyond what is necessary to provide the service and are not used for training. In addition, users should ensure that the tool they use provides the necessary rights for publication.
  • Prefer paid or enterprise versions over free tools. Many AI tools that are free to use are not private by default. We recommend checking whether your institution offers an enterprise version of a tool or a locally hosted model. Enterprise versions or locally hosted models are more likely to include privacy protection, although this is not guaranteed and should still be verified.