Redesigning Assessment for the Age of Agentic AI: A Real-World Reflection

September 23, 2026

This past summer (July and August here in Calgary), I taught EDER 619.26, Leadership for Learning: Policy, Governance, and Community, a fully online course in the University of Calgary’s Master of Education (MEd) program. Most of my students were K-12 teachers in Alberta pursuing advanced training while working full time. This was my first time teaching the course, and I used it as an opportunity to redesign my assessment approach in response to a concern I have followed closely in recent months: the threat that agentic AI poses to academic integrity in online courses.

I want to share what I did, why I did it, and how another instructor could try the same approach.

The Problem I Set Out to Solve

In online courses we have relied on the same assessment tools for decades: discussion board posts and a final paper submitted at the end of the term. These formats work well when the greatest risk to academic integrity is a student copying a classmate’s work. They work less well when a student can hand an assignment to an AI agent and receive a finished product in return. I wanted an assessment structure that valued process over product and made real-time, verifiable engagement central to how students earned their grades.

Photo by Canva Studio on Pexels.com

My Approach

I built the course as a hybrid model with mandatory video conference (i.e., Zoom) sessions, scheduled well in advance so students knew what to expect. I selected a small number of required readings and asked students to locate supplementary readings on their own through the library databases, connected to the weekly themes and the in-class tasks. I adopted a flipped classroom structure: students completed the readings before each session and arrived prepared to apply them.

Seventy-five percent of the course grade came from three real-time learning tasks tied to the Zoom sessions. For two of the three sessions, I brought a current Canadian policy or governance news story to the group, chosen because the course readings focused on the Canadian context. For the third, I built a synthetic educational case with the assistance of Claude, and I told my students I had done so. In each session, students worked in self-selected groups for a set period, initially ten minutes and later shortened to five, to apply the readings to the case or news story and produce a shared artifact documenting their thinking. They then uploaded that artifact to a course Dropbox.

I informed students from the outset that I was trying this format for the first time and that I expected to adjust it as I learned what worked. I asked for their feedback throughout the term and changed the process in response, including the shortened submission window after students reported that some groups kept working past the agreed time.

I also changed how I graded. I was not looking for a polished, consensus-driven product. I wanted evidence of student thinking, including open questions and points of disagreement that a group had not resolved. I told students that AI tool use was permitted under the University of Calgary Faculty of Graduate Studies Artificial Intelligence Guidelines and would not affect their grade either way. Most reported that they spent the bulk of their time in conversation with their groupmates rather than using AI tools, largely because the time constraint left little room for anything else.

The remaining twenty-five percent of the grade was a synthesis paper. Students combined the required and supplementary readings, the bibliographic sources they had shared with classmates during the sessions, and the two news stories and the synthetic case addressed in class into a single integrated paper.

Steps to Try This Assessment Approach

  1. Schedule your synchronous sessions early. Set the dates and times for all mandatory sessions at the start of the term and communicate the expectation of real-time participation.
  2. Curate a limited reading list. Choose a small set of required readings and ask students to locate supplementary sources through the library databases, tied to the weekly themes.
  3. Flip the classroom. Assign readings for completion before each session and communicate this expectation in writing.
  4. Prepare a case or current news story for each session. Select material relevant to your discipline and your students’ context, or construct a synthetic case with AI assistance. If you use AI assistance, disclose it to your students.
  5. Set a group task with a firm time limit. Give students 5 to 10 minutes to work in self-selected groups, apply the readings to the case, and produce a shared artifact documenting their discussion and reasoning.
  6. Collect the artifact through a shared drop point. I used the D2L / Brightspace Dropbox, but you could also use a shared document, or similar tool works well. Set the submission window based on your own testing; shorten it if students report that groups continue working past the agreed time.
  7. Grade for process, not polish. Communicate to students that you are evaluating evidence of learning and engagement with the readings, not a finished, consensus-driven product. Tell them unresolved questions and disagreements are acceptable and worth documenting.
  8. State your AI expectations explicitly. Clarify whether AI tool use is permitted for the in-class task and confirm that it will not affect grading either way, consistent with your institution’s guidelines (if they have them).
  9. Build a synthesis assignment. Ask students to integrate the required readings, their self-sourced supplementary readings, and the material from each session into a single paper at the end of the term.
  10. Offer an alternate path for students who miss a session. Schedule a makeup session or provide an equivalent assessment based on the same weekly reading and format.

What I Would Change Next Time

Students told me they would have benefited from more time to discuss each case or news story during the sessions. I plan to extend the collaboration window in future offerings of the course.

Reflection

Most of my students had not encountered agentic AI before this course and some did not know the term. That gap became a useful discussion point, even though artificial intelligence and academic integrity were not the stated focus of the course. Students reported that the sessions felt more purposeful than a standard discussion board, in part because they knew each session would produce a graded outcome. I plan to repeat this assessment structure. This updated assessment approach replaced a format that has gone unchanged in online learning for decades and shifted the emphasis toward collaboration, real-time problem solving, and process over product.

This was by no means a perfect experiment, but in the end, the effort was worth it… And I’ll close by saying that the students were — and are — brilliant, thoughtful, and inspiring.

___________

Share this post – Redesigning Assessment for the Age of Agentic AI: A Real-World Reflection – https://drsaraheaton.com/2026/09/23/redesigning-assessment-for-the-age-of-agentic-ai-a-real-world-reflection/

Sarah Elaine Eaton, PhD, is a Professor and Research Chair in the Werklund School of Education at the University of Calgary, Canada. Opinions are my own and do not represent those of my employer.

Leave a Comment » | artificial intelligence, education, higher ed | Tagged: , , , , , , , , , , , , , , | Permalink
Posted by Sarah Elaine Eaton, Ph.D.

The University of Toronto Settlement Is a Turning Point for Academic Integrity in Canada

May 22, 2026

On April 1, 2026, Justice Meaghan M. Conroy of the Federal Court of Canada issued a consent judgment confirming that the Easy EDU tutoring companies infringed the copyright of the University of Toronto and three named professors: Robert Gazzale, Lisa Kramer, and Ai Taniguchi. Joe Friesen covered the story for The Globe and Mail (paywalled, but worth tracking down). The settlement requires Easy EDU to pay $1 million in damages, plus HST and interest. A counterclaim by Easy EDU against the university was dismissed without costs. The case had been before the courts since 2022.

Friesen reported that Easy EDU reproduced course outlines, slide presentations, lecture notes, and assignments without authorization. In some instances, the company provided tests written by faculty with answers included, placing students at direct risk of academic misconduct violations. One published adjudication described how 180 students received a study package containing 22 questions that corresponded directly to questions a professor had written for an assessment. The student named in that case received a grade of zero and a 28-month suspension.

This case matters beyond copyright law and anyone who works in academic integrity in Canada should be paying attention Why? Because the case could set a precedent for how Canadian institutions respond when students use file-sharing services, term paper mills, or engage so-called academic consultants.

What the Settlement Exposes

Let’s be clear: the Easy EDU case did not emerge in isolation. The case reflects a pattern I and others have documented extensively. In Academic Integrity in Canada: An Enduring and Essential Challenge, which Julia Christensen Hughes and I edited and published with Springer in 2022, our contributors wrote about contract cheating in Canada in a number of chapters. Collectively, we traced the commercialization of academic support services as part of a broader commodification of higher education. The volume includes chapters on contract cheating in Canada and on EdTech-enabled contract cheating, both of which point to the same structural condition that made Easy EDU possible: a market for services that operates in parallel to formal education, targeting students under pressure, with few regulatory constraints.

Canada has no legislation against contract cheating companies. The U of T settlement is a copyright remedy, not a criminal one. Easy EDU was not shut down because what it did was illegal under a contract cheating statute. It was held responsible for reproducing materials that belonged to the university and its faculty. This distinction is important here because copyright law filled a gap that academic integrity policy could not. 

The Platformization of Academic Misconduct 

A recent 2026 study published in the British Educational Research Journal adds another dimension to this picture. My PhD student Gengyan Tang led this study, with Wei Cai and me as co-authors. Tang, Cai, and I examined commercial academic misconduct appeal services operating in China’s digital marketplace and found that these agencies target Chinese international students through platforms such as Xiaohongshu (RED). These services operate in a regulatory and ethical grey zone, packaging appeal support as a marketable product and translating case outcomes into metrics like success rates.

Tang, Cai, and I conceptualize this process as self-platformization: commercial actors reorganizing educational assistance in alignment with platform economies. The same logic applies to tutoring services like Easy EDU. These are not tutoring companies in any traditional sense. They are platform-aware businesses that use algorithms, social media, and scale to insert themselves into students’ academic lives at precisely the moments when students are most vulnerable.

The students in the U of T case were not, for the most part, bad actors. Many were international students navigating unfamiliar institutional systems, in some cases at risk of losing their study permits. Easy EDU identified that vulnerability and built a business model around it. One student, identified only by initials in the published adjudication, faced a 28-month suspension, not because they set out to deceive, but because a commercial service supplied unauthorized materials and they used them.

In our study, Tang, Cai, and I argue that institutions have invested heavily in prevention and detection but have largely ignored the post-violation stage. That gap is where commercial services can operate with relative ease because there is nothing stopping them from doing so. Universities focus on catching misconduct. Academic consulting services (i.e., contract cheating companies) profit from what happens next, whether that means supplying unauthorized test answers before an assessment or, as we found in our research, or coaching students through misconduct hearings afterward.

So What’s Next?

The U of T injunction permanently restrains Easy EDU from making further use of the university’s course materials. The university has committed to directing settlement proceeds toward student academic supports. These are constructive outcomes, and also insufficient on their own.

In Academic Integrity in Canada: An Enduring and Essential Challenge, I, together with other contributors, called for legislation that would deter contract cheating firms from operating in Canada. That call has gone largely unanswered. The U of T case demonstrates that copyright enforcement can achieve results where academic integrity policy alone cannot, but copyright litigation is expensive, slow, and available only to institutions with the resources to pursue it. The case took four years to resolve, an in the financial climate we are in today, many institutions simply cannot absorb that kind of cost.

What the settlement does accomplish is normative because it establishes, through a federal court consent judgment, that reproducing course materials for commercial tutoring purposes constitutes copyright infringement. It names the professors whose intellectual property was taken and affirms, as U of T vice-provost Heather Boon stated, in Friesen’s article, that faculty own the copyright in their course materials and the university will support them in protecting it. That is a meaningful public statement. It signals that institutions are prepared to act, and that the legal tools to do so exist.

The Bigger Picture

Academic integrity is not simply a student conduct problem, but rather a structural problem shaped by institutional design, assessment practice, resource inequity, and the commercialization of educational support. The Easy EDU case sits at the intersection of all of these.

The students who attended those tutoring sessions needed academic support. Easy EDU positioned itself as that help, at a price, with materials it had no right to distribute. The university’s commitment to redirecting settlement funds to student supports is the right response. It will not be enough without sustained investment and clearer procedural guidance for students facing misconduct allegations. Institutions across our country happily received international student tuition fees, on the assumption that students are admitted have the academic skills and preparation they need to succeed. By and large, we still tend to blame the students if they lack academic skills or knowledge of how to navigate the higher education system. When the students turn to third parties whom they believe can help them fill their skills gap, historically, it is the students who are held responsible while companies operating in the background simply line their pockets with profits without any repercussions. This is the first time, to my knowledge, that a commercial supplier of academic services operating in Canada has faced a monetary penalty for facilitating academic misconduct.

Better institutional supports, clearer procedural guidance for students facing misconduct allegations, and platform-aware integrity education are not peripheral concerns. They are the conditions under which commercial exploitation becomes less attractive. In our study, Tang, Cai, and I call for a post-violation framework that attends to digital infrastructures and addresses students during crisis moments, not only before them. I continue to believe that work is overdue.

Kudos to the team at U of T for pursuing this case. You’ve now set a precedent that others can follow.

References

Eaton, S. E., & Christensen Hughes, J. (Eds.). (2022). Academic integrity in Canada: An enduring and essential challenge. Springer. https://doi.org/10.1007/978-3-030-83255-1

Friesen, J. (2026, May 22). University of Toronto reaches settlement for $1-million in damages from tutoring company. The Globe and Mail. https://www.theglobeandmail.com/canada/article-university-of-toronto-reaches-settlement-1-million-damages-tutoring  

Tang, G., Eaton, S. E., & Cai, W. (2026). Academic misconduct appeal services in China: Platform logics, self-platformization and implications for integrity education. British Educational Research Journal. https://doi.org/10.1002/berj.70130

______

Share this post: The University of Toronto Settlement Is a Turning Point for Academic Integrity in Canada – https://drsaraheaton.com/2026/05/22/the-university-of-toronto-settlement-is-a-turning-point-for-academic-integrity-in-canada/

Sarah Elaine Eaton, PhD, is a Professor and Research Chair in the Werklund School of Education at the University of Calgary, Canada. Opinions are my own and do not represent those of my employer.

Comments Off on The University of Toronto Settlement Is a Turning Point for Academic Integrity in Canada | academic integrity, contract cheating, education, higher ed | Tagged: , , , , , , , , | Permalink
Posted by Sarah Elaine Eaton, Ph.D.

Interfacing with the Future: Reflections on the National Day of Learning 2026

April 1, 2026

On March 28, 2026, I had the pleasure of joining educators from across Canada for the National Day of Learning, hosted by Let’s Talk Science. This one-day, nation-wide professional learning event brought together K–12 teachers, post-secondary educators, and policy leaders to explore some of the most pressing issues shaping education today, with artificial intelligence high on the agenda.

I was invited to deliver a session titled “Interfacing with the Future: Wearable AI and Academic Integrity for K–12 and Higher Ed.” What follows are a few reflections and key ideas from that conversation, hosted by Dr. Alec Couros.

Moving into the Postplagiarism Era

One of the central ideas framing my talk is postplagiarism. In this reality, artificial intelligence is no longer an external tool that students occasionally use, but rather, it is embedded into everyday life and learning.

Students are already engaging with AI in ways that challenge traditional notions of authorship, originality, and academic work. The question is no longer if students will use AI, but how.

This shift requires a corresponding change in how we think about academic integrity. Detection and surveillance, long relied upon as primary strategies, are no longer sufficient. Instead, we must rethink how we design learning environments that foster integrity from the ground up.

From Tools to Wearables: How AI is Advancing

A key focus of my presentation was the rapid evolution from AI tools to AI wearables — particularly smart glasses and other forms of cosmetically invisible interfaces. The talk was based, in part, on our recent article in Canadian Perspectives on Academic Integrity. 

Wearable technologies integrate AI directly into our physical experience of the world. Rather than pulling out a device, users can access real-time information, transcription, and prompts seamlessly through their field of vision.

This shift introduces both opportunities and tensions:

  • Cognitive offloading: Learners can reduce mental load by accessing information instantly. (Phill Dawson has done some great work on cognitive offloading that I recommend reading.)
  • Enhanced presence: Wearables allow users to maintain eye contact and engagement without device distraction.
  • Efficiency gains: Tasks such as note-taking or translation can be automated in real time.

At the same time, these benefits come with real challenges including information overload, privacy concerns, and technical limitations. More importantly for educators, they fundamentally disrupt assumptions about what it means to “know” something independently.

New Technology ≠ Cheating

One of the most important messages I emphasized is this: new technology does not automatically equal academic misconduct.

If a tool is permitted, then its use is not cheating. The real issue lies in unauthorized use or misuse in ways that create unfair advantage. 

We must also remain attentive to equity and accessibility. Some wearable technologies may be used as accommodations, making it essential that our integrity policies are inclusive and nuanced rather than rigid and punitive.

Designing for Integrity (Not Surveillance)

Rather than doubling down on detection, I encourage educators to shift their focus toward designing for integrity.

This means:

  • Prioritizing assessment validity: If an AI system can complete a task without genuine understanding, then the task itself needs to be rethought.
  • Moving beyond “gotcha” approaches: Surveillance-based strategies erode trust and are increasingly ineffective.
  • Supporting diverse learners: Students bring different technological access, needs, and experiences. Our designs must reflect that.
  • Building a culture of integrity: Integrity is not enforced; it is cultivated through meaningful learning experiences.

Bridging K–12 and Post-Secondary Education

Another key theme was the gap between K–12 and post-secondary expectations.

In K–12 environments, students are often encouraged to explore technology as part of their learning. In contrast, post-secondary institutions frequently operate under the assumption that students already understand complex academic integrity rules.

As AI continues to evolve, this gap becomes more pronounced. We need stronger alignment across educational sectors to ensure that students are supported, rather than being set up for failure, as they transition between systems. (Myke Healy has a great paper on the topic of GenAI in the K-12 context that is worth reading.) 

Looking Ahead

If there is one takeaway from this experience, it is this: wearable AI is not a future scenario. It is already here.

As educators, we are being called to respond not with fear, but with thoughtful, research-informed approaches. The challenge is not simply to manage technology, but to reimagine teaching, learning, and assessment in ways that remain meaningful in an AI-integrated world.

Events like the National Day of Learning remind me of the power of community. Bringing educators together to share ideas, ask difficult questions, and explore new possibilities is essential as we navigate this rapidly changing landscape.

Thank you to Let’s Talk Science and to Dr. Alec Couros for the opportunity to be part of this important conversation, and to all the educators who continue to lead with curiosity, courage, and care.

______________

Share this post: Interfacing with the Future: Reflections on the National Day of Learning 2026 –  https://drsaraheaton.com/2026/04/01/interfacing-with-the-future-reflections-on-the-national-day-of-learning-2026/

Sarah Elaine Eaton, PhD, is a Professor and Research Chair in the Werklund School of Education at the University of Calgary, Canada. Opinions are my own and do not represent those of my employer.

Comments Off on Interfacing with the Future: Reflections on the National Day of Learning 2026 | education, technology | Tagged: , , , , , , , , , , , , , , , , , , , , , | Permalink
Posted by Sarah Elaine Eaton, Ph.D.

Call for Proposals: Special issue on Postplagiarism and Generativism: Human-AI Hybrid Approaches to Ethical Teaching, Learning, and Assessment

March 17, 2026

Special Issue Call for Papers

Postplagiarism and Generativism: Human-AI Hybrid Approaches to Ethical Teaching, Learning, and Assessment

For publication in the Journal of University Teaching and Learning Practice

Guest editors

Background

Every new technology brings with it societal and moral panic (Orben, 2020). When the Internet first became popular, concerns about plagiarism increased. Even though there is scant empirical evidence that the Internet was actually responsible for increases in rates of plagiarism, the perception that new technology resulted in more academic cheating persisted (Panning Davies & Howard, 2016).

Some plagiarism scholars have been emphatic that the majority of student plagiarism cases are not an intent to deceive, but rather a lack of academic literacy and poor academic practice, and have even advocated for disposing of plagiarism in academic misconduct policies in favour of increased student support (Howard, 1992; Jamieson & Howard, 2021). The idea that plagiarism could be decoupled from academic misconduct seems somewhat unlikely, but by the 2020s it was obvious to some that generative artificial intelligence (GenAI) would have an impact on writing, and by extension, on plagiarism (Mindzak & Eaton, 2021).

In response to these technological shifts, various frameworks have emerged to conceptualize academic integrity in the GenAI era. The postplagiarism framework, first introduced by Eaton (2021, 2023) and since discussed by scholars worldwide (Bali, 2023; Bagenal, 2024; Kenny, 2024), offers one approach. Other perspectives, such as Generativism (Pratschke, 2023), AI Literacy frameworks (Ng et al., 2021; Pretorius & Cahusac de Caux, 2024), and UNESCO’s Guidance for Generative AI in Education (2023), provide complementary or alternative viewpoints on similar phenomena.

Postplagiarism is based on six tenets (Eaton, 2023): (1) human-AI hybrid writing will become the norm; (2) creativity can be enhanced by AI; (3) AI can help to overcome language barriers; (4) we can outsource control of our writing to AI, but we do not outsource responsibility for what is written; (5) attribution remains important; and (6) historical definitions of plagiarism may require rethinking.

Empirical testing of these and related frameworks has shown differing levels of acceptance and application across educational contexts (Kumar, 2025).

Equity, Diversity, Inclusion, and Accessibility in a Postplagiarism Age

As higher education institutions aim to promote social justice through equity, diversity, and inclusion (EDI), GenAI holds the potential to either break down or reinforce barriers related to linguistic, cultural, socioeconomic, and ability differences requires critical examination.

Assessment practices should be designed proactively to enable all students to demonstrate their learning without being unfairly disadvantaged by their personal characteristics or circumstances (Tai et al., 2022). Similarly, McDermott (2024) highlights the importance of considering accessibility, equity, and inclusion in assessment and academic integrity.

GenAI offers opportunities to enhance equity by providing personalized support, overcoming language barriers, and assisting learners with diverse needs. However, without careful implementation, it may exacerbate existing inequities through unequal access to technology, algorithmic biases, or assessment designs that privilege certain ways of knowing and communicating.

In this special edition, we propose to examine the broader question: “How are pedagogies, learning, and teaching approaches evolving in response to GenAI, and what frameworks best support ethical academic practice in a postplagiarism landscape?”

We invite researchers and practitioners to submit their original research papers exploring the transformation of teaching, learning, and assessment in a GenAI age. We welcome both theoretical and empirical contributions, including positions that may present contrasting viewpoints. Potential topics of interest include, but are not limited to:

  • New developments in postplagiarism, generativism, and other emerging frameworks for understanding academic integrity in the GenAI era
  • Empirical studies testing these frameworks in different contexts and disciplines
  • The use of these frameworks to design or reform academic misconduct policies and procedures
  • The relationship between GenAI, academic literacies, and related competencies (e.g., digital literacy, information literacy)
  • Pedagogical approaches that embrace GenAI while maintaining academic integrity
  • Case studies of successful integration of GenAI into teaching, learning, and assessment
  • Critical perspectives on the limitations or challenges of current approaches to GenAI in education
  • Position papers presenting new or alternative frameworks for understanding GenAI in teaching and learning

We particularly encourage submissions that engage in dialogue with existing frameworks, offering either supportive evidence or critical alternatives. Our goal is to foster a robust debate about the future of teaching and learning in a GenAI (and even a post-GenAI) world.

We welcome submissions from both established researchers and early-career scholars from diverse academic and cultural backgrounds. All submissions will be peer-reviewed by an international panel of experts. Accepted papers will be published in a special issue of the Journal of University Teaching and Learning Practice.

Types of publications accepted into this Special Issue

The types of publications that are eligible for acceptance into this Special Issue include:

  • Research papers
  • Review articles (e.g., systematic review or meta-analysis)
  • Case studies and evidence-based good practice examples

Developing a high-quality proposal

We recommend the creation of a single document in Word (.doc or .docx) format that contains the following:

  • Proposed article title
  • Proposed authors names, affiliations, and ORCid
  • A clear evidence-based rationale for the line of inquiry proposed
  • Research question(s)
  • Proposed method (for both theoretical and empirical manuscripts)
  • Practice-based implications of the proposed research

The word limit for the proposal is 250 words (not including references) and is designed to give the Editorial Team a sense of the rigour of the manuscript proposed and the possible implications of such research. The Editorial Team may return with an invitation to combine similar manuscripts. Acceptance of proposals does not guarantee acceptance of final manuscripts.

Timeline

  • Proposals due – April 30, 2026
  • Proposal acceptance notifications: May 14, 2026
  • Full articles due: August 31, 2026

Submit your abstract via this online form: https://forms.gle/6sKjc2jkKGWCtGgw7

For further information contact Professor Sarah Elaine Eaton, University of Calgary.

References

Bali, M. (2023, March 3). Are We Approaching a Postplagiarism Era? https://blog.mahabali.me/educational-technology-2/are-we-approaching-a-postplagiarism-era/

Bagenal, J. (2024). Generative artificial intelligence and scientific publishing: Urgent questions, difficult answers. The Lancet, 403(10432), 1118–1120. https://doi.org/10.1016/S0140-6736(24)00416-1

Eaton, S. E. (2021). Plagiarism in Higher Education: Tackling Tough Topics in Academic Integrity. Bloomsbury.

Eaton, S. E. (2023). Postplagiarism: Transdisciplinary ethics and integrity in the age of artificial intelligence and neurotechnology. International Journal for Educational Integrity, 19(1), 1–10. https://doi.org/10.1007/s40979-023-00144-1

Orben, A. (2020). The Sisyphean cycle of technology panics. Perspectives on Psychological Science, 15(5), 1143–1157. https://doi.org/10.1177/1745691620919372

Howard, R. M. (1992). A plagiarism pentimento. Journal of Teaching Writing, 11(2), 233–245.

Comments Off on Call for Proposals: Special issue on Postplagiarism and Generativism: Human-AI Hybrid Approaches to Ethical Teaching, Learning, and Assessment | academic integrity, education, postplagiarism | Tagged: , , , , , , , , | Permalink
Posted by Sarah Elaine Eaton, Ph.D.

What Should We Be Assessing in a World with AI? Insights from Higher Education Educators

November 25, 2025

The arrival of generative AI tools such as ChatGPT has disrupted how we think about assessment in higher education. As educators, we’re facing a critical question: What should we actually be assessing when students have access to these powerful tools?

Our recent study explored how 28 Canadian higher education educators are navigating this challenge. Through in-depth interviews, we discovered that educators are positioning themselves as “stewards of learning with integrity” – carefully drawing boundaries between acceptable and unacceptable uses of chatbots in student assessments.

Screenshot of an academic journal article header from Assessment & Evaluation in Higher Education, published by Routledge. The article title reads: “What should we be assessing exactly? Higher education staff narratives on gen AI integration of assessment in a postplagiarism era.” Authors listed are Sarah Elaine Eaton, Beatriz Antonieta Moya Figueroa, Brenda McDermott, Rahul Kumar, Robert Brennan, and Jason Wiens, with institutional affiliations including University of Calgary, Pontificia Universidad Católica de Chile, Brock University, and others. The DOI link is visible at the top: https://doi.org/10.1080/02602938.2025.2587246.

Where Educators Found Common Ground

Across disciplines, participants agreed that prompting skills and critical thinking are appropriate to assess with chatbot integration. Prompting requires students to demonstrate foundational knowledge, clear communication skills, and ethical principles like transparency and respect. Critical thinking assessments can leverage chatbots’ current limitations – their unreliable arguments, weak fact-checking, and inability to explain reasoning – positioning students as evaluators of AI-generated content.

The Nuanced Territory of Writing Assessment

Writing skills proved far more controversial. Educators accepted chatbot use for brainstorming (generating initial ideas) and editing (grammar checking after independent writing), but only under specific conditions: students must voice their own ideas, complete the core writing independently, and critically evaluate any AI suggestions.

Notably absent from discussions was the composition phase – the actual process of developing and organizing original arguments. This silence suggests educators view composition as distinctly human cognitive work that should remain student-generated, even as peripheral tasks might accommodate technological assistance.

Broader Concerns

Participants raised important challenges beyond specific skill assessments: language standardization that erases student voice, potential for overreliance on AI, blurred authorship boundaries, and untraceable forms of academic misconduct. Many emphasized that students training to become professional communicators shouldn’t rely on AI for core writing tasks.

Moving Forward

Our findings suggest that ethical AI integration in assessment requires more than policies, it demands ongoing conversations about what makes learning authentic in technology-mediated environments. Educators need support in identifying which ‘cognitive offloads’ are appropriate, understanding how AI works, and building students’ evaluative judgment skills.

The key insight? Assessment in the AI era isn’t about banning technology, but about distinguishing between tasks where AI can enhance learning and those where independent human cognition remains essential. As one participant reflected: we must continue asking ourselves, “What should we be assessing exactly?”

The postplagiarism era requires us to protect academic standards while preparing students for technology-rich professional environments – a delicate balance that demands ongoing dialogue, flexibility, and our commitment to learning and student success.

Read the full article: https://doi.org/10.1080/02602938.2025.2587246

______________

Share this post: What Should We Be Assessing in a World with AI? Insights from Higher Education Educators – https://drsaraheaton.com/2025/11/25/what-should-we-be-assessing-in-a-world-with-ai-insights-from-higher-education-educators/

Sarah Elaine Eaton, PhD, is a Professor and Research Chair in the Werklund School of Education at the University of Calgary, Canada. Opinions are my own and do not represent those of my employer.

Comments Off on What Should We Be Assessing in a World with AI? Insights from Higher Education Educators | academic integrity, artificial intelligence, education, higher ed | Tagged: , , , , , | Permalink
Posted by Sarah Elaine Eaton, Ph.D.

« Previous Entries