Guidelines for Use of Generative AI in Research Grant Applications
Grant and fellowship applications are extensions of our scholarly work and should uphold the same standards of accuracy, integrity, and ethical practice expected in peer-reviewed publications. As generative AI (GenAI) tools evolve, the following recommendations provide preliminary guidance for their responsible use in preparing research proposals.
1. Review Sponsor Guidance
Consult funding agency policies, if any, regarding GenAI use in proposal development and peer review. Agencies, such as NSF, and many private funders, such as Spencer Foundation, have issued notices on AI use.
2. Verify All AI‑Generated Information
GenAI tools should not be treated as reliable factual sources. Always verify AI‑generated content using credible sources such as Tutt Library or by consulting a librarian.
3. Protect Data and Maintain Security
GenAI tools may introduce privacy or security risks. When working with sensitive or internal materials, log into the College’s secure Microsoft Copilot “AI sandbox,” which prevents data from being used to train public AI models.
4. Use GenAI Ethically and Transparently
Except for basic functions (spelling, grammar, translation), any substantive use of GenAI must be disclosed – typically in a Methods or Acknowledgements section – and cited appropriately in APA, MLA or Chicago style.
5. Assign Authorship Responsibly
Organizations including COPE, WAME, and JAMA Network state that AI tools cannot be credited as authors because they cannot take responsibility for submitted work.
6. Fulfill Author Responsibilities
Authors remain responsible for all AI‑generated or AI‑assisted content. AI use must be disclosed, including the tool name, model/version, manufacturer, and content produced. AI‑generated images are discouraged unless part of formal research methods.
Conclusion
WAME (2023) identifies core principles for responsible scholarship:
- Only humans can be authors.
- Authors must acknowledge their sources.
- Authors must take public responsibility for their work.
- Editors and reviewers should disclose AI use.
- Editors need tools to address AI‑related challenges in publishing.