✍️By Kennedy O’Neil from Encode Canada.
Kennedy is an undergraduate sociology student at McGill University, with a minor in psychology, and a writer for Encode Canada.
📌 Editor’s Note: This piece is part of our Recess series, featuring university students from Encode’s Canadian chapter at McGill University. The series aims to share insights from university students on current issues in AI ethics. In this article, Kennedy O’Neil shares students’ first-hand experiences of the confusing nature of AI guidelines at universities, and the roles students play in shaping how they are improved.
Photo credit: UX Indonesia on Unsplash.
“There’s a big difference between using AI to generate something and then claiming it as your own versus using it to supplement something you are already doing,” a second-year neuroscience student at McGill University told me. Like many students, they have been left to define their own boundaries around AI use. Since 2021, students have faced a new and largely uncharted dilemma in education: how to engage responsibly with artificial intelligence. As these tools become increasingly integrated into academic work, educators and institutions alike have struggled to adapt. The absence of clear, consistent, and enforceable standards on generative AI in universities has shifted the responsibility for ethical decision-making onto students, intensifying moral stress and competitive pressure.
At universities, students can easily leverage GenAI chatbots such as ChatGPT, Gemini, and Copilot, which have transformed the possibilities for engagement with academic material. The expansive capabilities of such programs can facilitate the efficient summarization of texts, extraction of key themes, and the simplification of information. For learners who struggle with traditional instructional methods (e.g. independent readings and auditory-based lectures), AI can serve as a supplementary teaching assistant by presenting information in a more accessible manner, through the aforementioned capabilities. For these reasons, many students feel justified in turning to AI to break down challenging content and to keep pace with heavy course loads. A third-year student studying history told me, “There is no way I would be able to do all of my course readings independently. I have to use AI to keep up,” implying that she feels using AI is not optional. Though sources report varying estimates of the percentage of students who admit to using AI regularly, a recent KPMG Canada report found that 73 per cent of Canadian students use generative AI tools in their schoolwork.
While institutional guidance does exist, university AI policies are often inconsistent across courses and generally easy to circumvent, shifting the burden of ethical decision-making onto students. Across my own five courses this semester, policies varied widely: three professors prohibited GenAI in the submission of written assignments, one detailed that AI is not to be used for any aspect of the course, and another did not mention it in the syllabus at all. This inconsistency leaves students, like myself, to interpret the standards on our own. If it’s acceptable to use AI to draft outlines for one course, why should it be unacceptable in another? Additionally, it is often clear that professors have no way of deciphering if AI has been used, particularly in the context of using GenAI to study. One student I spoke with detailed how insignificant AI policies can feel in practice, stating that they have “never been caught” for GenAI use, and that many students “AI check themselves” using online detection programs prior to submission, in order to avoid penalization.
As students struggle to individually decipher how much AI use is “too much,” many report feelings of guilt for using AI to assist them with schoolwork (Jo, 2026), while others express concern that they will be placed at an academic disadvantage if they do not utilize AI tools (Kanabar, 2023). I spoke with a sociology and communications student who was told by a classmate that “everyone’s gonna use AI for this assignment,” which “really disappointed (her.)” She notes that she was “working on it slowly, actually reading through the articles”, while it seemed like her classmates were flying through with the help of AI. Eventually, she began to worry about the evaluation of her work and whether or not she was at a disadvantage, asking “How does my work compare to one that’s made by a computer?” In addition to these comparisons, the student expressed an overarching concern that when she uses AI, she “feel(s) like (she) is not developing (her) skills.” When regulation of AI use is not effective or consistent, students can feel pressured to use these tools despite personal discomfort.
This ambiguity does not simply create confusion; it adds another source of stress to an already high-pressure environment. University is a markedly stressful time for young people. Studies show that there has been a significant increase in reports of high stress levels and mental health challenges experienced by Canadian university students in recent years (Linden, Boyes, and Stuart 2021). From global crises to social media and rapidly advancing technology, students are experiencing education in novel ways that can evoke stress. In addition to the other challenges that university students face, they must now determine where to draw the line between using AI as a legitimate learning tool, and outsourcing their work to AI, thereby committing plagiarism. When expectations remain unclear, decisions around AI become a recurring source of anxiety, as students are constantly self-evaluating their AI use.
While universities around the world continue to implement AI policies, moral dilemmas surrounding AI are evolving faster than regulations can address them. In the absence of effective institutional guidance, students are increasingly left to mitigate these dilemmas together. Creating space for students to discuss AI use with peers can foster a sense of community and may reduce the isolation that many students feel around making these decisions. Additionally, student-organized and led workshops on ethical AI use can improve visibility around AI in academic settings while mitigating feelings of shame or fear of repercussions. One example is an education initiative led by Encode Canada. Through workshops with local high schools in Montreal, Encode’s team is working to bring ethical AI use to the forefront of classroom discussion, fostering curiosity, collaboration, and brainstorming for how to address the issue at its roots. Intervention and education at the high school level are crucial to students’ success in university and beyond, as the high school years serve as a period of immense development of critical and creative thinking skills. Ultimately, it is up to us, as students, not only to decide how we individually use AI, but also to define what ethical and effective learning looks like.
References
Higgs, J.M. and Stornaiuolo, A. (2024), Being Human in the Age of Generative AI: Young People’s Ethical Concerns about Writing and Living with Machines. Read Res Q, 59: 632-650. https://doi.org/10.1002/rrq.552
Clare Baek, Tamara Tate, Mark Warschauer, “ChatGPT seems too good to be true”: College students’ use and perceptions of generative AI, Computers and Education: Artificial Intelligence, Volume 7, 2024, 100294, ISSN 2666-920X, https://doi.org/10.1016/j.caeai.2024.100294.
Fang, S., Barker, E., Arasaratnam, G., Lane, V., Rabinovich, D., Panaccio, A., O’Connor, R.M., Nguyen, C.T. and Doucerain, M.M. (2025), Resilience, Stress, and Mental Health Among University Students: A Test of the Resilience Portfolio Model. Stress and Health, 41: e3508. https://doi.org/10.1002/smi.3508
Bell, Russell, Between Prohibition and Practice: Institutional AI Policy Contradictions (September 28, 2025). Available at SSRN: https://ssrn.com/abstract=5541138 or http://dx.doi.org/10.2139/ssrn.5541138
Linden B, Boyes R, Stuart H. Cross-sectional trend analysis of the NCHA II survey data on Canadian post-secondary student mental health and wellbeing from 2013 to 2019. BMC Public Health. 2021 Mar 25;21(1):590. doi: 10.1186/s12889-021-10622-1.
Jo, H. (2026). The impact of guilt on student interactions with generative AI technology. Ethics & Behavior, 36(1), 13–39. https://doi.org/10.1080/10508422.2025.2466152
Kanabar, V. (2023). An Empirical Study of Student Perceptions When Using ChatGPT in Academic Assignments. In: Zlateva, T., Tuparov, G. (eds) Computer Science and Education in Computer Science. CSECS 2023. Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering, vol 514. Springer, Cham. https://doi.org/10.1007/978-3-031-44668-9_30
Butalid, R., Wright, C. & Kherroubi Garcia, I. (Eds.). (2025). The State of AI Ethics Report (Volume 7) – AI at the Crossroads: A Practitioner’s Guide to Community-Centered Solutions. Montreal AI Ethics Institute. DOI: 10.5281/zenodo.17328882. Available at: https://montrealethics.ai/state
Johnston, H., Wells, R.F., Shanks, E.M. et al. Student perspectives on the use of generative artificial intelligence technologies in higher education. Int J Educ Integr 20, 2 (2024). https://doi.org/10.1007/s40979-024-00149-4
Yun Dai, Why students use or not use generative AI: Student conceptions, concerns, and implications for engineering education, Digital Engineering, Volume 4, 2025, 100019, ISSN 2950-550X, https://doi.org/10.1016/j.dte.2024.100019.
Saraf, Kapil (2024) Learning with artificial intelligence: how students decide and then use AI. Masters thesis, Concordia University. Text
Vieriu, A. M., & Petrea, G. (2025). The Impact of Artificial Intelligence (AI) on Students’ Academic Development. Education Sciences, 15(3), 343. https://doi.org/10.3390/educsci15030343
Sun, R. C., & Hui, E. K. (2012). Cognitive competence as a positive youth development construct: a conceptual review. TheScientificWorldJournal, 2012, 210953. https://doi.org/10.1100/2012/210953
Valentyna Nechyporenko, Nataliia Hordiienko, Olena Pozdniakova, Ellina Pozdniakova-Kyrbiatieva, Yuliya Siliavina. How often do University Students use Artificial Intelligence in Their Studies?. WSEAS Transactions on Information Science and Applications. 2025;22:203-214. https://doi.org/10.37394/23209.2025.22.18
Kar SK, Bansal T, Modi S, Singh A. How Sensitive Are the Free AI-detector Tools in Detecting AI-generated Texts? A Comparison of Popular AI-detector Tools. Indian Journal of Psychological Medicine. 2025;47(3):275-278. doi:10.1177/02537176241247934