Following a recent Advance HE conference presentation, Dr Paul Jones, Associate Dean for Education and Student Experience, Aston University, reflects on how artificial intelligence is doing more than transforming higher education. It is exposing the long-held assumptions about learning, assessment and knowledge that have quietly shaped the sector for generations.

We often talk about artificial intelligence as if it represents a revolution in higher education. In some ways, it does. It changes what students can produce. Yet there is another way to understand what is happening. 

We are witnessing an AI relevation: a moment in which artificial intelligence reveals the traditional learning script that higher education has carried for generations. 

Technology as pressure test 

Education has experienced technological disruption before. The printing press transformed access to knowledge and unsettled authority. Calculators challenged assumptions about mathematical competence. Wikipedia forced educators to rethink expertise, reliability and evaluation. In each case, the technology did not simply create a new problem, it pressure-tested existing educational scripts. 

AI is doing something similar, but it is pressing on deeper foundations. It is not only challenging how students access knowledge or perform particular tasks. It is forcing us to ask what we mean by learning, originality, understanding and academic integrity. 

AI as postmodern provocateur 

This is where I find a postmodern lens useful. Postmodernism is often caricatured as a rejection of truth, evidence or standards. Used carefully, it encourages us to examine the narratives and power structures that shape what counts as knowledge. It asks who defines the rules, whose ways of knowing are privileged, and what becomes invisible when one version of education is treated as the norm. 

In that sense, AI has become an unexpected postmodern provocateur. It is not postmodern by design. It has no philosophical intention. But its presence prompts the kinds of questions postmodern thinkers have long asked. Why do we treat certain forms of assessment as more legitimate than others? Why do we equate a polished final product with learning? Who benefits from our current definitions of originality, authorship and criticality? What forms of thinking remain hidden because they do not fit existing assessment structures? 

Beyond detection 

These questions matter because much of our response to AI has focused on command and control. Institutions have understandably looked to policy, detection and guidance. Those things have their place. Students and staff need clarity. Academic integrity matters. But if our main response is to build a more heavily surveilled version of the old university, we miss the deeper opportunity. 

AI is showing us that we have over-relied on the finished output. The essay, report or exam answer has often stood in for learning itself. Yet learning is so much more than the final artefact. It includes uncertainty, judgement, revision, dialogue, rejection, interpretation and the gradual construction of meaning. AI did not create that complexity. It has simply made it harder to ignore. 

This shifts the challenge for educators. Instead of asking only whether students used AI, we might ask whether they can explain how they used it, why they trusted or rejected particular outputs, and how their own thinking changed through the process. The assessable object can move from the polished answer alone to the reasoning behind it. 

That does not mean abandoning standards. It means becoming clearer about what our standards are for. If we value critical thinking, we need to design opportunities for students to show it. If we value judgement, we need to ask students to justify decisions. If we value originality, we need to clarify whether originality lies only in unaided production or also in framing problems, selecting evidence, challenging assumptions and making defensible interpretations. 

Making thinking visible 

A practical way forward is to build AI use around three principles: transparency, judgement and accountability. 

Transparency means students should be clear about where and how AI has contributed to their work. Judgement means they should be able to critique AI outputs rather than simply accept them. Accountability means they remain responsible for the final decisions, interpretations and claims they make. 

These principles move us beyond permitted or prohibited use. They invite students into a more mature relationship with technology, one based not on concealment but on explanation. They also give educators a richer basis for assessment. Rather than trying to detect whether AI has been used, we can ask students to make their thinking visible. 

This could involve reasoning maps that chart key decisions, AI process logs that show how prompts and responses developed, workflow narratives that account for revision and rejection, or critical comparisons between human and AI interpretations. None of these approaches treat AI as a shortcut to be feared or a magic solution to be celebrated. They treat it as part of a learning process that needs to be examined. 

The deeper point is that AI asks us to revisit the purpose of higher education itself. If technology can produce fluent answers, then perhaps our role is not simply to train students to produce answers. It is to help them ask better questions, recognise weak arguments, work with uncertainty, make ethical decisions and explain the reasoning behind their choices. 

That is why the postmodern provocateur idea matters. AI is not just another tool in the educational toolkit. It is a mirror held up to the traditional script. It asks us to look again at practices that had become familiar enough to feel natural. It reminds us that our systems of learning and assessment are not inevitable. They were designed, inherited, normalised and maintained. That means they can also be reimagined. 

The opportunity, then, is not simply to respond to AI. It is to use AI as a catalyst for more honest conversations about education. What do we really value? What kinds of learners do our systems privilege? What forms of thinking are made visible, and what remains hidden? 

AI may well be revolutionary. But its most important contribution may be revelatory. It reveals the cracks, tensions and inherited scripts that were already present. Now we need to decide what to do with that knowledge. 

Perhaps the future of AI in higher education depends less on whether we can control the technology, and more on whether we are willing to let it challenge us. 

What assumptions about learning has AI encouraged you to revisit in your own practice?