When did you first hear about ChatGPT? For most of us, it was in 2022 when GenAI entered our lives. It immediately triggered excitement and anxiety in the professional world, especially in higher education, which was expected to take the lead in the conversation. Thousands of posts and articles began to appear, discussing the impact of GenAI on the future of universities, skills, learning, and thinking.
The critiques continue to circulate, but a growing stream of research is actually going in a different direction: GenAI is not a threat to learning – it is an opportunity to bridge the gap and change what was already fragile and outdated in higher education, including course assessments and activities which struggle to remain meaningful in AI-rich environments. In other words, yes, now is the right time to adapt; it is the right time to rethink how we design courses, set learning goals, and assess and teach students to think.
Why is This Change Happening Now?
If AI can produce acceptable academic output, what learning objectives should we be trying to achieve?
AI can perform many course tasks, yet many courses continue to rely on assessments of individual or group work that assume independent authorship (Lubbe et al., 2025). But is it really possible to know? This assumption of independence becomes increasingly unrealistic as students routinely use GenAI to take notes, summarize readings, develop basic structures, or even create complete reports. The evidence shows that students use AI because it increases their efficiency, clarity of writing, and task performance, especially in demanding academic work when they are juggling multiple projects and deadlines (Nakatani & Jiang, 2025).
Of course, students are very creative in finding the best ways to earn good grades. And if the output to be evaluated is not AI-resistant, what are we actually assessing? Unfortunately, this reflects an apparent misalignment between assessment and learning goals vs. contemporary learning conditions. Students use AI strategically and instrumentally, driven by performance pressure (Al-Mamary, 2025). If AI can complete all the work required for a course, what do students learn? AI did not break or destroy education; it simply exposed weaknesses that have been there for a long time: slow adaptation to change.
What Skills Students Need
In 2019, a few years before the era of massive GenAI adoption, the OECD published the Learning Compass report, which shifted the focus from "How do we stop AI?" to "What will learning look like in 2030 and beyond?"
Students first need to build strong foundational skills, including cognitive, health, and social and emotional foundations, as they are essential prerequisites for all further learning across the curriculum. In this context, higher education becomes the place where the foundational skills are further developed and where social and interpersonal skills are cultivated through communication, collaboration, conflict resolution, and respectful engagement with diverse perspectives. Even if AI is around, these skills remain essential.
These foundations form the basis for the transformative competencies that can be the most distinctive in the future:
- creating new value through innovation, creativity, entrepreneurship;
- navigating tensions and dilemmas while balancing competing interests, using ethical reasoning, applying systems thinking;
-
taking responsibility by acting ethically and with the long term in mind.
Education is the key to developing these competencies, which are essential for success in our new world.
Finally, there is one more set of future skills under the "must have" category - those that enable individuals to shape their own learning across a lifetime. These skills include goal setting, self-direction, reflection, self-evaluation, agency, and the capacity to work productively with others. In the age of AI, education should go beyond teaching content and help students learn how to learn, act, and continue learning.
Human-Centered Pedagogy in an AI Age
One potential strategy: to turn courses into labs for critical thinking and experimentation. They can function as spaces for education based on self-regulated learning, experiential learning, and collaboration – where AI potential is leveraged and knowledge development is guided within human-centered pedagogies (Muthukrishna et al., 2025).
Prohibition-based approaches to AI in higher education are often ineffective. Students are skeptical of AI bans, as the business world follows an absolutely different path. Instead of prohibiting new technologies, maybe we have a unique opportunity to rethink what we teach and how we assess the learning outcomes, paying more attention to the learning process itself: reasoning, interpretation, judgement, decision-making, and reflection (Yang et al., 2024). Instead of focusing primarily on final outputs, which can now be generated by AI, course assignments and assessment systems can be redesigned to demand deeper understanding and creativity – not just finding quick answers – with critical thinking becoming a central element.
For instance, students can work on tasks that require critique and comparison, where they can refute AI-generated responses or contextualize them by providing their justified opinion and documenting the interaction. In addition, AI has immense potential for personalization in learning, which we have yet to leverage in higher educational institutions.
When AI use is made explicit and assessable, students can engage with it more responsibly and use AI as a scaffold rather than a substitute for thinking. This is where we see higher-order learning outcomes: when decision-making and metacognition is prioritized over production.
How AI is Changing Higher Education
In this context, as AI capabilities continue to evolve, what should we do?
As one expert noticed, “the bar of expectation should go up.” To maintain meaningful learning in a world where AI can perform almost any tasks, we need to raise the cognitive demands of our courses – and the human side in our teaching.
During a focus group discussion with educational professionals, which we run within the Mind The Gap (MTG) project, several moves were highlighted that could make learning more resilient to the AI era and more challenging to game. One approach is to redesign the questions we ask so that they can’t be simply answered by AI, for example, by swapping some parameters, changing contexts, or altering tasks to specific situations that AI can’t “know”. Another move is to deliberately force thinking by developing more customized versions of GenAI for our courses, where students can engage in learning dialogue with AI. They can clarify tasks, reason, ask questions, reflect – in other words, actively learn with the help of AI rather than outsource their cognitive work to it.
We also need to protect fundamental knowledge and keep retrieval practices alive, since memorization remains a necessary component of deep learning. But the global role of educators is shifting. Today, they act more and more like an orchestra conductor who brings together multiple activities, tools, experiences, standards, and assignments into a coherent learning composition. When all these elements come together, they create a relevant learning journey that stays with learners for years. The goal – and challenge here – is to prepare future professionals who know how to leverage AI and do work that it cannot replace.
What Comes Next
The changes we are going through signal to us that it's time to stop preserving pre-AI models of education and start exploring new strategies. We can redesign courses and redesign assessments around transparency, reasoning, reflection, ethical judgement – and to shift the focus from originality to accountability, from polished outputs to processes and reasoning, from surveillance to pedagogy. One initiative in this direction is the MTG project aimed at developing a pilot platform that can help educators redesign courses to ensure the course’s relevance in the AI world.
In the process of transformation and adaptation, institutional support is also essential. Adoption, clarity on the rules, and professional development – all matter. And building AI literacy among educators and students is a necessary foundation today (Honigsberg et al., 2025). AI is not going away. The real question is how educational institutions will evolve with it and use it to define the next era of learning.
References
Al-Mamary, Y. H. (2025). A comprehensive model for AI adoption: Analysing key characteristics affecting user attitudes, intentions, and use of ChatGPT in education. Human Systems Management, 44(6), 978–999. https://doi.org/10.1177/01672533251340523
Honigsberg, S., Watkowski, L., & Drechsler, A. (2025). Generative Artificial Intelligence in Higher Education: Mediating Learning for Literacy Development. Communications of the Association for Information Systems, 56, 35. https://doi.org/10.17705/1CAIS.05640
Lubbe, A., Marais, E., & Kruger, D. (2025). Cultivating independent thinkers: The triad of artificial intelligence, Bloom’s taxonomy and critical thinking in assessment pedagogy. Education and Information Technologies, 30(12), 17589–17622. https://doi.org/10.1007/s10639-025-13476-x
Muthukrishna, M., Dai, J., Panizo Madrid, D., Sabherwal, R., Vanoppen, K., & Yao, H. (2025). AI Can Revolutionise Education but Technology Is Not Enough: Human Development Meets Cultural Evolution. Journal of Human Development and Capabilities, 26(3), 482–492. https://doi.org/10.1080/19452829.2025.2517740
Nakatani, K., & Jiang, Y. (2025). Understanding business students’ GenAI usage and perception and associated implications. Industry and Higher Education, 09504222251388167. https://doi.org/10.1177/09504222251388167
Yang, Y., Luo, J., Yang, M., Yang, R., & Chen, J. (2024). From surface to deep learning approaches with Generative AI in higher education: An analytical framework of student agency. Studies in Higher Education, 49(5), 817–830. https://doi.org/10.1080/03075079.2024.2327003
Written by
