This policy applies to:

It is crucial that the use of AI/LLM tools does not breach the University of Ottawa (UofO) academic integrity policies and guidelines, as well as any other related policies. These tools should be used to support, not replace, a person’s skills and ideas.
Graduate students and faculty supervisors must uphold the highest standards of academic quality and research integrity. This requires two‐fold transparency:

1. A pre-approval process that guarantees stakeholders understand how the AI/LLM tool will be utilised.
2. Provide clear disclosure to audiences, including a full description and proper citation of any use of an AI/LLM tool.

Pre-Approval Authorities

Use of AI/LLM tools may be pre-approved by one or more of the following:

• Trainee supervisor(s) (and, when appropriate, faculty collaborators) when using AI/LLM tools to design, create, or present research components (e.g. code, audio, images, videos) and outputs required for the degree (e.g., TAC reports, research project reports, posters, oral presentations, honours theses, graduate theses, annual progress reports).
• Research Ethics Board (REB) for AI/LLM use involving animal or human research, including the storage, analysis, or processing of related data.
• It is recommended that AI/LLM policies be included in course syllabi, Brightspace sites, or via written approval from course coordinators, faculty, or instructors. These policies will be created by course coordinators in collaboration with all instructors involved. If no specific policy exists for class work, the use of generative AI or consultation will be considered similar to peer assistance.
• Approval from the course coordinator is required for the use of AI/LLMsupport in grading coursework, assignments, or exams by teaching assistants, grading assistants, or instructors. It is recommended to use uOttawa approved AI platforms such as Copilot.
• Supervisors and TAC members, specifically for research output (e.g. code, audio, images, video) and thesis writing and editing. Students must request permission at the time they receive approval to write their thesis. This request—and any subsequent requests during writing—must align with the University of Ottawa’s Academic Regulation C-7 and Ontario’s graduate-thesis requirements,10 which emphasize original research and independent capabilities. The student’s request and approval should include a summary of:
  • AI tool(s) name/version (e.g., ChatGPT 4.0)
  • Intended uses (e.g., grammar edits; idea generation; drafting text; figure creation)
  • Approximate AI-like generated content (e.g., “5–10% of draft text or editing”)

Note: Supervisors, course coordinators, instructors, members of the Thesis Advisory Committee (TAC), and other academic stakeholders are authorized to prohibit, in whole or in part, the use AI/LLM technologies for any course requirements or for any research activities associated with a thesis or dissertation.

AI Acknowledgment and Appropriate Citations

Trainees must acknowledge and cite any generative AI assistance in all coursework and research activities. When in doubt, default to full disclosure using one or more of these methods:

• Include an AI Acknowledgments section with a heading such as “Use of Generative AI” or “AI Usage” in areas like the Preface, Acknowledgments, Methods section, or after your reference list. Briefly describe which tool(s) and version you used, when, and for what purpose.
• In-Text Citations Treat AI output as personal communication or software. Example (APA-style): “…as refined through AI-assisted phrasing (ChatGPT, “prompt”, June 15, 2025).” If quoting AI verbatim, use quotation marks and cite similarly.
• Reference-List Entry Follow your discipline’s conventions. Example (APA): OpenAI. (2025, June 15). ChatGPT (Version 4.0) [Large language model].
• In the case of using generative AI such as ChatGPT for producing written content, the entered prompt must be included.

Example #1

Student-A has been struggling with lab work for their thesis, so they have had no time to do an  extensive literature search on one of their main projects. Student-A’s thesis supervisor has requested a concise introductory page for the project within 10 days. Student-A knows that there is no way they can deliver something in that timeframe.

Example #2

Student-B has just completed a comprehensive set of Western Blot experiments using blood samples from a 4-month-long experiment. However, in several of the gels, the control group did not run properly, and there seems to be a problem with protein loading consistency. Student-B knows that without these controls, the entire 4-month experiment is in jeopardy. 

References

1. Should researchers use AI to write papers? Group aims for community-driven standards, (2024).
2. Buriak, J. M. et al. Best Practices for Using AI When Writing Scientific Manuscripts. ACS Nano 17, 4091-4093 (2023).
3. Chetwynd, E. Ethical Use of Artificial Intelligence for Scientific Writing: Current Trends. Journal of Human Lactation 40, 211-215, (2024).
4. Cooperman, S. R. & Brandão, R. A. AI assistance with scientific writing: Possibilities, pitfalls, and ethical considerations. Foot & Ankle Surgery: Techniques, Reports & Cases4, 100350, (2024).5 Khalifa, M. & Albadawy, M. Using artificial intelligence in academic writing and research: An essential productivity tool. Computer Methods and Programs in Biomedicine Update
5. Using artificial intelligence in academic writing and research: An essential productivity tool 100145, (2024).
6. Peters, U. & Chin-Yee, B. Generalization bias in large language model summarization of scientific research. Royal Society Open Science 12, 241776, (2025).
7. Wang, J. & Fan, W. The effect of ChatGPT on students’ learning performance, learning perception, and higher-order thinking: insights from a meta-analysis. Humanities and Social Sciences Communications 12, 621, (2025).
8. Is it OK for AI to write science papers? Nature survey shows researchers are split, (2025).