AI Causing 'Moral Injury' to Lecturers Policing Its Use in Canadian Universities

The Rise of AI in Canadian Classrooms and Its Hidden Toll

Generative artificial intelligence (AI) tools like ChatGPT have revolutionized higher education across Canada, offering students unprecedented access to instant content generation. However, this shift has placed university lecturers in a precarious position, tasked with upholding academic integrity amid widespread student use. Recent research from Trent University highlights a profound issue: the 'moral injury' experienced by instructors as they police AI misuse. This emotional and cognitive strain stems from constant vigilance, unreliable detection methods, and a sense of futility in preserving original thought in assignments.

In Canadian universities from Ontario to British Columbia, lecturers report redesigning entire courses—swapping digital essays for handwritten exams—to combat AI reliance. At Trent University, English literature instructor Mac Fenwick noted that three-quarters of his first-year students used AI for term papers last year, prompting a return to pen-and-paper assessments. Students who aced AI-assisted coursework often failed in-person exams, revealing a gap in genuine cognitive engagement.

Defining Moral Injury in the Context of AI Enforcement

Moral injury occurs when individuals are forced into actions that violate their core ethical beliefs, leading to deep psychological distress. For Canadian lecturers, this manifests as the anguish of accusing students of cheating based on flawed AI detectors or altering teaching methods to 'AI-proof' assessments. Trent University's study, led by Amanda Paxton, coordinator of journalism and creative writing at the Durham campus, details how Ontario writing instructors face heightened cognitive and emotional labor. They grapple with spotting AI-generated text, confronting students, and questioning the purpose of their role amid technological disruption.

Paxton explains that instructors feel overwhelmed, frustrated, and ashamed when they miss AI use, compounded by a lack of institutional guidance. This echoes findings from the University of Windsor, where secondary educators report similar burdens, though higher education faces amplified stakes with research and credentialing.

Lecturer Experiences: Frontline Stories from Canadian Campuses

Across Canada, lecturers share harrowing accounts. At the University of Waterloo, Fraser Easton mandates handwritten assignments to foster cognition and deter shortcuts. 'It's like going to the gym and asking someone else to lift the weights,' says Trent's Andrew Monti, who teaches AI ethics in communication courses. He distinguishes productive AI use for data analysis from bypassing learning.

In Calgary, professors note a spike in AI-aided plagiarism, while Thompson Rivers University (TRU) reports rising academic integrity breaches linked to generative AI. Concordia University logged over 100 AI misconduct cases in the 2025-2026 year alone. These pressures contribute to burnout, with instructors redesigning syllabi multiple times annually.

Statistics Revealing the Scale of AI Use and Cheating

A KPMG Canada report reveals 73 percent of students rely on generative AI for schoolwork, with 57 percent feeling they are cheating and 54 percent fearing detection. Nearly half view it as their first instinct for tasks. At TRU, AI-related violations surged, mirroring national trends where 58 percent use AI as a tutor, 48 percent for research, and 38 percent questionably.

• 73% of Canadian students use AI regularly for assignments (KPMG, 2025).
• 57% admit it feels like cheating.
• Over 100 AI misconduct cases at Concordia in one year.
• Spike in plagiarism at Calgary post-secondaries.

These figures underscore why lecturers feel like they are in an unwinnable 'arms race,' as Fenwick describes.

Challenges with AI Detection Tools and Their Fallout

AI detectors like Turnitin promise salvation but deliver false positives, flagging human work as machine-generated and eroding trust. Canadian instructors avoid them due to unreliability, fearing wrongful accusations that amplify moral injury. University Affairs warns that detection focus creates workload stress from student appeals.

Without reliable tools, lecturers resort to surveillance software for exams or in-class writing, further straining resources. Paxton calls for federally vetted detectors to ease the burden on underfunded teaching centers.

University Adaptations: From Essays to In-Class Writing

Canadian institutions are innovating. UBC and Waterloo discourage detectors, favoring disclosure policies where students must note AI use. Queen's University emphasizes original work, while USask holds students accountable for knowing rules. Many, like Trent and Paxton's courses, mandate handwritten submissions to ensure authenticity.

Process-based assessments—focusing on drafts and reflections—gain traction, rewarding development over final products. However, these shifts demand time instructors lack, exacerbating overload.

Student Perspectives: Reliance, Guilt, and Skill Erosion

Students acknowledge AI's allure for efficiency but worry about critical thinking atrophy. Fenwick observes AI-dependent learners falter without tools, degrading cognitive skills. A KPMG survey shows half fear detection, yet usage persists, highlighting a disconnect between policy and practice.

Institutional and Provincial Policy Responses

Ontario universities lead with AI hubs and guidelines, but a patchwork persists. HEQCO urges ethical, responsive policies balancing integrity and innovation. U15 institutions grapple with GenAI integration, prioritizing fairness and accessibility. Provinces like Saskatchewan allow ethical AI, but federal coordination lags.

The Federal Role: Upcoming AI Strategy and Recommendations

Minister Evan Solomon's office recognizes AI's classroom transformation, pledging support for digital literacy and integrity without 'policing.' Paxton urges federal funding for teaching centers and reliable detectors. Experts like York University's Aimi Hamraie advocate campus limits on corporate AI tools to curb normalization.

Broader Implications for Canadian Higher Education

Unchecked AI risks widening inequality, as low-income students may overuse it due to pressures. Lecturers' moral injury signals systemic strain, potentially driving talent away. Positive shifts include AI literacy courses, preparing graduates for an AI economy.

Pathways Forward: Solutions and Optimism

Solutions blend pedagogy and policy: AI-resilient designs, ethical training, collaborative forums, and national guidelines. Trent's Provost AI Working Group builds capacity for safe use. By addressing moral injury, Canada can harness AI's potential while safeguarding education's soul.

Balanced approaches empower lecturers, foster student responsibility, and maintain integrity. As Monti notes, AI is a tool, not a replacement for human intellect.

Photo by Maxime Doré on Unsplash