A new measure of curricular exposure to large language models is constructed by combining task-level estimates of LLM capabilities with course descriptions from more than 1,000 U.S. colleges and universities, showing that colleges have recognized the instructional challenge posed by generative AI but have made limited observable changes to how student learning is assessed.
The rapid adoption of artificial intelligence (AI), particularly large language models (LLMs), has fundamentally disrupted how learning is demonstrated and evaluated in higher education. Tasks that once served as proxies for understanding-such as writing essays, solving problem sets, or producing computer code-can now be generated superficially by AI systems with minimal human effort. This paradigm shift raises a critical ethical question: how should learning be evaluated when traditional indicators of competence are easily outsourced? This paper examines the ethical challenges of educational evaluation in the age of AI from a university-level perspective. We argue that the core problem extends beyond academic dishonesty to a deeper misalignment between assessment practices and the learning outcomes they are intended to measure. Evaluation regimes that rely on artificial constraints risk measuring compliance, access, or concealment rather than genuine understanding, reasoning, or judgment. By analyzing institutional responses and presenting empirical survey data, we highlight the need for alternative assessment models that emphasize process over product. The goal is to establish ethically informed assessment strategies that preserve student agency and accountability in an automated age.
Md Zarzees Uddin Shah Chowdhury, Samin Khan· 0 citations
The literature examining the impact of artificial intelligence (AI) on higher education often boils down to the question of whether AI will replace teachers. This study examines the issue more broadly — through the lens of the diverse nature of academic tasks. A multidimensional analysis is conducted to show how academic work is being redistributed among tasks that AI can replace, augment, or cannot meaningfully displace. The study uses qualitative content analysis on a purposive corpus of 46 documents, including scientific papers, reports, policy and practitioner texts, and higher education commentary published between 2020 and 2026. A 3×3 analytical matrix was developed, reflecting the degree of distribution of functions between humans and AI ("Replace," "Augment," or "Human-dominant") across three basic dimensions of academic activity (teaching, research, and socio-technical work). The findings show that AI is strongest in codifiable and routine tasks such as basic educational content generation, objective grading, literature processing, drafting, and administrative support. By contrast, tasks involving judgment, ethics, interpretation, mentoring, relationship-building, assessment design, governance, and public trust remain human-dominant. The research concludes that AI does not eliminate academic work; it reorders it, shifting value from content transmission toward learning design, interpretive expertise, and institutional stewardship.
Amir Ghorbani, M. Blankesteijn· Foresight and STI Governance· 0 citations
The rapid advancement of Artificial Intelligence (AI) has led to its widespread adoption in highereducation. A growing body of research has examined its educational implications, includingeffects on learning and academic integrity. Building on this literature, this paper investigateshow students’ orientations toward learning, effort, and ethics relate to their use of AI inacademic contexts. The study was conducted on a sample of 115 undergraduate and graduatebusiness students using a structured questionnaire. Data were analyzed using Chi-squaretests and Cramer’s V coefficients. The results reveal significant relationships betweenstudents’ orientations and AI use. Students who prioritize efficiency, minimal effort, and finaloutcomes are more likely to use AI frequently, trust its outputs without verification, and rely onit for assignments. Likewise, students with more tolerant attitudes toward unethical academicbehavior are more likely to use AI as a shortcut in academic tasks. In contrast, students whovalue the learning process, theoretical knowledge, and academic integrity are more likely tocritically evaluate AI-generated content and consult additional sources. The findings suggestthat AI acts primarily as an amplifying tool rather than an independent driver of academicbehavior, reflecting pre-existing attitudes toward learning and ethics.
Aleksandar Vučković, Ernest Vlačić, A. Davidovic· Notitia· 0 citations
It is found that there is no significant differential effect of GenAI availability on grades overall or among previously lower-performing students, and the findings temper concerns that GenAI inflates grades and reduces students's satisfaction.
J. Dumlao, Meng Wang, Zhonghan Xie et al.· 0 citations
Generative Artificial Intelligence has had a profound and continuing impact on both research and on higher education in recent years. There are a number of issues arising from this which are particularly relevant to teaching research methods at undergraduate level. Generative AI has considerable potential to facilitate research, notably in reviewing literature, and as such its use should arguably be part of the syllabus for students on a taught degree course. However much of the dialogue, so far, about generative AI in the context of university teaching, particularly at the undergraduate level, has focused on concerns that its use by students weakens established forms of assessment and could inhibit learning through being used as a short cut to obtaining a degree. To complicate the picture further, undergraduates are preparing to enter an uncertain environment for employment, with a concern that generative AI will directly affect many graduate-level jobs and that during their careers it will change the landscape within which they work in ways that cannot possibly be predicted. This paper builds on earlier conceptual work presented at ECRM in 2025 and uses data gathered in a focus group of recent Business Management graduates who took a research methods in their final year. It uses their experience to identify pointers for how use of generative AI can be incorporated into the teaching of research methods and how this can prepare a forthcoming generation of students for the workplace. The recent graduates participated in a focus group in which they reported varying experiences of use of AI whilst they were studying and following university. There were considerable differences between individuals in their propensity to engage with AI and in the resources that they used to support this engagement. However, a common theme highlighted by participants was the scope for universities to provide more thorough guidance about the practicalities of students’ use of AI.
Martin Rich, Amit Rawal· European Conference on Resea...· 0 citations
Generative artificial intelligence (AI) has entered everyday undergraduate study faster than the evidence base has kept up with it, and the evidence that does exist points in opposite directions. This paper synthesizes peer-reviewed and carefully delimited contextual research on how generative AI use relates to undergraduate academic performance, with attention to what Canadian universities can already act on. Searches of Google Scholar, Scopus, Web of Science, ERIC, and ScienceDirect covering 2022 to 2026 produced a two-tier evidence base: Tier 1 peer-reviewed empirical and synthetic studies of student learning outcomes, and Tier 2 supplementary sources admitted under stated justifications, including contextual Canadian and international surveys, one secondary-school field experiment, and one preprint mechanistic study. Experimental syntheses report medium to large short-term gains when ChatGPT is built into instruction. Survey work points the other way: frequent unstructured use tracks with procrastination, self-reported memory problems, and slightly lower grades, and unrestricted access during practice has been shown to depress later unaided performance. Purpose of use reconciles most of that disagreement, because a tool that scaffolds thinking behaves very differently from one that replaces it. Canadian peer-reviewed studies document heavy campus adoption and considerable student ambivalence about integrity and learning, yet almost none link purpose-differentiated use to measured performance. Closing that gap matters for assessment redesign, AI-literacy programming, and the credibility of the credentials Canadian universities issue.
Pragalvha Sharma· Canadian Journal of Business...· 0 citations