August
04
AI isn’t the problem. Our understanding of learning might be.
Over the past year or so, one conversation has become increasingly familiar in higher education.
Academic colleagues are expressing genuine concern about the growing use of artificial intelligence by students. Questions about plagiarism, academic integrity and authenticity dominate staff meetings, conferences and online discussions. The concern is understandable. If AI can produce essays, reports and presentations in seconds, what exactly are universities assessing?
Perhaps that is the wrong question, and the more important question might be this:
What should universities develop in students when AI becomes a normal part of professional life?
The age of knowledge scarcity is over
Universities were built during a period when knowledge was relatively scarce. Expertise resided in books, journals and lecturers. The role of education was to help students acquire that knowledge, apply it and demonstrate their understanding through assessment.
Artificial intelligence changes that equation. Knowledge is now abundant – you can’t imagine how threatening that is to someone who spent a lifetime building that specialist knowledge.
Information can be retrieved, summarised and explained in seconds. AI can compare theories, generate examples and even draft coherent arguments. Competing with AI on information retrieval is becoming increasingly difficult, and arguably unnecessary.
What remains scarce is something far more valuable: Judgement.
Perhaps this is where higher education should refocus its attention. It is the same refocusing this series has argued marketing leadership needs, moved from one profession to another. An education system that measures itself by how much a student can recall is an extraction model: it depletes the appearance of learning without building the capability that actually compounds over a career. A system that measures itself by how well a student thinks, decides, acts and reflects is an investment model, and the distinction matters more now than it ever has, because AI has made the extraction version trivially easy to counterfeit.
Building on established educational thinking
The ideas explored here are not an attempt to replace established educational theory. Rather, they build upon a long tradition of research into how people learn and how professionals develop expertise.
More than eighty years ago, John Dewey argued that education should move beyond the transmission of knowledge towards learning through experience and reflection. Later, David Kolb’s Experiential Learning Cycle demonstrated how experience, reflection, conceptual understanding and experimentation combine to create continuous learning. Donald Schön extended this thinking through his concept of the reflective practitioner, highlighting the importance of reflecting both during and after professional action.
At the same time, Bloom’s Taxonomy challenged educators to move beyond simple recall towards higher-order cognitive skills such as analysing, evaluating and creating. More recently, John Biggs’ principle of constructive alignment reinforced the importance of ensuring that learning outcomes, teaching activities and assessment all support the capabilities universities seek to develop.
These ideas remain highly relevant.
What has changed is the context in which they are applied.
Artificial intelligence has dramatically reduced the effort required to retrieve information, summarise research and generate coherent first drafts. As a result, the competitive advantage of graduates is shifting away from knowledge retrieval and towards judgement, critical thinking and responsible decision-making.
This is not the first time education has adapted to technological change. The printing press reduced the need to memorise texts. Calculators shifted mathematics towards problem-solving rather than arithmetic. Search engines transformed access to information. AI represents the next stage in that evolution.
The challenge for universities is therefore not to compete with artificial intelligence at tasks it performs exceptionally well. Instead, it is to focus on developing the capabilities that remain distinctly human: the ability to think critically, exercise sound judgement, act responsibly and continually learn through reflection.
The capability framework proposed in this article is intended as a practical response to that challenge. It draws upon established educational principles while placing professional judgement and human–AI collaboration at the centre of learning and assessment.
From knowledge acquisition to capability development
Rather than viewing education as the transfer of knowledge, universities might begin to view it as the development of professional capability.
One way of describing that capability is through a simple learning flywheel. This is the working draft. The full model is illustrated at the end of the blog.
Unlike a linear process, a flywheel represents continuous improvement. Each cycle strengthens the next.
Professional practice begins with understanding. Students should learn to define problems clearly, gather evidence, challenge assumptions and explore multiple possibilities before rushing towards solutions. AI can help expand thinking, but it cannot determine which questions are the most important to ask.
Information alone has little value until choices are made. Students need to exercise judgement, evaluate alternatives, weigh trade-offs and justify their decisions using evidence rather than intuition alone. AI can generate options. Professionals remain responsible for choosing between them.
Ideas only create value when they are implemented. Graduates must be capable of communicating decisions, leading implementation, adapting to change and turning strategy into meaningful action. Execution remains the bridge between thinking and impact.
This may be the most overlooked stage of learning. Reflection transforms experience into expertise. By evaluating what worked, what failed and what should change next time, students continually refine their judgement. Reflection feeds directly back into better thinking, allowing the flywheel to gather momentum throughout a professional career.
The decision engine inside the flywheel
The flywheel describes how capability develops. It still leaves an important question unanswered.
How do professionals make better decisions?
One possible answer lies in a simple decision framework built around four questions, and it is worth working through what each one actually demands rather than treating it as a checklist. The first question is what we know: the facts of the situation, gathered honestly rather than selectively, and distinguished clearly from assumption. The second is what is shaping the situation: the forces acting on a decision that the facts alone do not reveal, whether that is organisational politics, market timing or constraint. The third is what matters most: the values, personal and organisational, that decide which trade-offs are acceptable and which are not, because facts and forces alone never produce a decision on their own. And the fourth is the decision itself: given everything above, what should we do, and can that choice be defended on the basis of the reasoning that produced it rather than on instinct alone.
This framework acts as the cognitive engine within the learning flywheel. Thinking is informed by facts and an understanding of external forces. Decisions are guided by organisational and personal values before choices are made. Action tests those choices in the real world. Reflection then generates new facts, reveals emerging forces, challenges existing values and improves future choices. Each rotation of the flywheel strengthens professional judgement.
What does this mean for assessment?
If universities continue to assess primarily the quality of a finished report, AI will continue to disrupt traditional approaches.
Perhaps assessment should focus less on the final product and more on the quality of the thinking that produced it. That means asking a different set of questions of a student’s work than the ones most assessment currently asks. Not simply whether the report is accurate and well written, but whether the student understood the problem before attempting to solve it. Not simply whether the recommendation is plausible, but whether it is justified, with the reasoning and the trade-offs made visible rather than hidden behind a confident conclusion. Not simply whether the plan was followed, but whether it was applied with the judgement to adapt when circumstances changed. And not simply whether the assignment is finished, but what the student can demonstrate they learned in producing it, including where their own thinking diverged from what the AI first offered them.
The use of AI would no longer be the central issue. The quality of judgement would.
Students could demonstrate how AI contributed to their thinking, where they challenged its suggestions and how their own professional reasoning shaped the final outcome. In doing so, assessment would move closer to the realities of contemporary professional practice, where nobody is marked down for using the tools available to them, and everybody is judged on what they did with what those tools gave them.
A different conversation
Artificial intelligence is undoubtedly changing higher education. However, it may also be revealing something more important, and something considerably closer to home for those of us who teach marketing specifically.
The instinct to catch students out is the instinct of a system that still believes its primary job is verifying what students know. It is an extraction posture applied to education: it treats the relationship between institution and student as one to be policed rather than developed, and it will lose an arms race against tools that will only get better at producing convincing outputs. The alternative is not to relax the standard. It is to change what the standard is measuring.
Perhaps universities have spent too much time assessing what students can remember and too little time developing how they think. (Actually, I think that is unfair to many gifted and focused lecturers I know who think and work at a deeper cognition activating level.) It is hard to ignore the reality that graduates who will thrive in an AI-enabled world are unlikely to be those who can reproduce the most information from memory. They will be those who consistently demonstrate the ability to think critically, decide wisely, act effectively and reflect continuously.
AI is not redefining the purpose of higher education. It is revealing it. In a world where knowledge is increasingly abundant, the true value of a university education lies in developing graduates who can think critically, decide wisely, act effectively and reflect continuously.
Here’s my attempt, standing on the shoulders of giants, at a working full flywheel model for module content and assessment structure designed to help lecturers, students and assessors focus on what matters most: valuable life-changing student experiences and lasting outcomes.
Let me know what you think. It will help me make better decisions, act more effectively and give me something new to reflect on.
Dewey (1938) Bloom et al. (1956; revised taxonomy by Anderson & Krathwohl, 2001) Kolb (1984) Schön (1983) Biggs and Tang (2011)
Discover more from jam partnership
Subscribe to get the latest posts sent to your email.
