Dissertation-Level AI Policy – Canadian Universities Catching Up?
The PhD and dissertation level is where AI policy gets most complicated and where I’ve seen the least institutional clarity.
I’ve been surveying graduate coordinators at 8 Canadian universities as part of an ongoing research project. Here’s what the data shows:
Only 2 of 8 had explicit PhD-level AI policies as of May 2026. The other 6 have policies that cover undergraduate work and mention AI but don’t address the specific challenges of dissertation research.
The dissertation-specific challenges nobody’s addressing:
– AI literature review assistance vs AI writing assistance (very different)
– AI data analysis tools in quantitative research (standard practice in many fields)
– AI transcription and qualitative coding (growing use in qualitative research)
– AI writing assistance for non-native English PhD students
These are real questions that graduate students are navigating without institutional guidance. They’re making their own decisions about what’s acceptable because their supervisors often don’t know either.
Anyone supervising graduate students – what are you telling them?
3 Replies
Join the discussion.
Log In to Reply
supervising my first master's student this year and the AI disclosure conversation was genuinely uncomfortable because I didn't have clear institutional guidance to give her. told her to disclose any AI use in her methodology section and we'd figure out the standards together. not ideal but honest.
the dissertation context also involves committee members with very different individual stances. a student can navigate their supervisor's AI policy and then face a completely different standard from an external examiner. until policies are applied at the program level with explicit committee-wide standards, the inconsistency will persist regardless of what the institution officially says.
The literature review distinction is the key one that gets missed. Using AI to summarize 200 papers is categorically different from using AI to write your theoretical framework. One is a research efficiency tool that's been part of academia in different forms for decades. The other is substituting AI analysis for the scholar's own synthesis. Graduate policies need to draw these lines explicitly.