When AI makes answers cheap, what are business schools for?

As I have outlined in a previous Think article, AI is making fluent analysis cheap and traditional signals of competence noisier. The question now is not only what business schools should teach in response, but what kind of value they themselves can credibly add.

The danger is not that serious education could disappear. It is that some of the outputs around which education and knowledge work have long been organised – the polished memo, the competent analysis, the fluent presentation, the confident synthesis – are easier to imitate. When that happens, the question is not whether the best human judgement remains better than AI. Often it does. The question is where real scarcity now lies, and whether our teaching, assessment and credentials still recognise it.

Business schools have already gone through two major shifts in their purpose and model. The first business schools were trade schools, created to professionalise management, to provide a language for business administration, and to help firms develop managers for increasingly complex organisations. The second epoch was the research transformation: the effort to give management education intellectual discipline, analytical depth and stronger foundations in economics, statistics, behavioural science and the social sciences.

From the 1960s onwards, business schools became more academically rigorous, more discipline-based and more analytically sophisticated. They also helped popularise managerial and financial frameworks, and spawned a new category of workers – analysts and professional experts. That brought benefits – it improved the quality of thought and created knowledge that has shaped how firms, markets and organisations are understood. But it also had costs. It centred on analysis, models and frameworks. It rewarded academic production in ways that did not always align with the most urgent questions facing organisations, and this often went unnoticed because polished frameworks had an aura of exclusivity.

AI may mark the beginning of a third epoch. The more fluent the surrounding discourse becomes, the more valuable it is to have institutions that can ask what is actually known, what is merely asserted, what mechanisms are at work, what evidence would change our view, and what follows for action. Business schools may need to return to their focus on helping run businesses. This is a demand for a more exacting form of business education, one that is more rigorous because the world is noisier, and more relevant because rigour that avoids the real questions will not do.

Business schools are not called to add a technology layer to an otherwise unchanged model, or to teach managers to become technologists. They need to help managers understand how AI alters the structure of work, the boundaries of firms, the location of expertise, the allocation of accountability, the nature of competition and the possibilities for new organisational forms, and equip them to navigate change.

The classroom as the test case

Case teaching is one of the great inventions of management education. By looking at a real-world management problem, participants are forced to take a position with incomplete information, listen to others, revise their views and confront the fact that such problems do not arrive neatly labelled by discipline. It asks people to move between finance, organisation, strategy, power, incentives and implementation, often without the comfort of a single correct answer. A good case discussion is a rehearsal for judgement.

But case teaching does not always achieve this. Cases can be too clean, too retrospective and too dependent on the professor’s choreography. Students may learn the performance of participation rather than the discipline of decision. They may learn to recognise the cues of a case discussion rather than develop the capacity to act when the cues are absent. This remains valuable, but it is not the same as standing on one’s feet in a live setting, where facts are incomplete, stakeholders are present and the implications of choice are still unfolding.

AI makes this distinction more visible. If the old case summary, the standard diagnosis and the elegant list of alternatives can be produced instantly, the classroom has to move further towards what cannot be so easily produced: the contested process of judgement. That means paying closer attention to how students frame problems, challenge evidence, handle disagreement, revise assumptions and move from insight to action.

The language of experiential learning is often used too loosely to refer to anything more active than a lecture: simulations, fieldwork, live cases, role plays, entrepreneurial experiments. Serious experiential learning requires participants to confront uncertainty, make choices, receive feedback and understand consequences. It should develop judgement under conditions closer to the world our students and executives actually inhabit.

AI can help here, by making learning not easier but more demanding. Students can use AI to generate alternative interpretations, interrogate assumptions, construct scenarios, simulate stakeholder responses, compare strategic options and test the coherence of a plan. Executives can bring live problems from their organisations and use AI not as an answer machine, but as a way of exposing what they have failed to consider.

The danger is that AI becomes another device for producing confident superficiality. This is why pedagogical design matters. The purpose is not to use AI because it is novel. The purpose is to use it to make weak thinking visible.

What business schools must now clarify

Business schools need to do more than “teach AI”: they must treat AI as a managerial, organisational and strategic phenomenon. Students need enough technical literacy to avoid naivety, but the central questions concern work, value, accountability, competition and institutional change.

That means assessment must change too. Schools should not only ask students to produce answers; they should ask them to frame problems, use AI intelligently, test assumptions, defend choices, persuade others and reflect on consequences. If competent-looking analysis can be generated on demand, the test must move closer to judgement, standards and responsibility.

Much of the knowledge about AI is being created outside universities: in technology companies, consultancies, start-ups, public agencies, banks, hospitals and elsewhere. No school can internalise all of this, but a good school can curate it. It can test external developments against evidence, theory and managerial experience. It can help distinguish demonstration from deployment, productivity from advantage, automation from transformation, and plausible claims from robust findings.

This curatorial role is valuable because the AI conversation is so noisy. There is hype from vendors, anxiety from incumbents, opportunism from advisers, overconfidence from technologists and defensiveness from professionals. Business schools are not immune from any of this. They are part of the same economy of claims, reputations and incentives. But at their best, they can create a more disciplined arena in which the conversation becomes less breathless and more useful.

The implications for students are both sobering and encouraging. Some capabilities that have traditionally signalled elite managerial potential are becoming easier to imitate. The person who can synthesise obvious material, present confidently and reproduce the language of strategy will be less differentiated than before. But the deeper capabilities remain valuable, and may become more valuable precisely because imitation is easier.

AI can help students prepare better, think more broadly, test themselves more rigorously and engage with a wider range of possibilities. But it cannot give them seriousness of purpose. It cannot make them care about consequences. It cannot substitute for the work of becoming someone who can exercise judgement in the presence of uncertainty, disagreement and pressure.

For faculty, the challenge is equally direct. We will need to be less attached to our role as providers of content and more committed to the design of serious learning. This is not a demotion, it is a more demanding form of authority. It requires us to bring research closer to action without reducing it to anecdote; to engage with technology without pretending to be technologists; and to work with practice without becoming consultants in academic dress. It also requires some humility, because AI will expose formulaic teaching just as surely as it exposes formulaic student work.

What AI reveals about knowledge institutions

When analysis, language and synthesis become cheaper, organisations have to ask where their real value lies. This applies to any institution whose authority depends on expertise, judgement, trust and the ability to turn knowledge into consequential action. If the visible output becomes easier to imitate, the basis of differentiation must move elsewhere. It lies in knowing not only what can be said, but what should be trusted; not only what options exist, but which ones matter; not only what a system can generate, but what an organisation can actually do.

This is where business schools have a broader role to play. At their best, they are not simply providers of content or credentials. They are institutions that can help society understand how management, technology, markets and organisations are changing. They can convene researchers, executives, entrepreneurs, regulators and students in a setting where claims are tested, assumptions are challenged and fashionable answers are slowed down long enough to be examined.

That role becomes more important, not less, in an AI-rich world. The more abundant fluent output becomes, the more valuable it is to have institutions capable of disciplined judgement. The more claims are made about productivity, transformation and disruption, the more important it is to distinguish adoption from impact, automation from value creation, and impressive demonstrations from organisational change.

For business schools, then, AI is not just a curriculum challenge. It is an institutional test. It asks whether we are producing articulate observers of business or people capable of acting with judgement. It asks whether we are teaching content or developing capacity. It asks whether our classrooms are places where uncertainty is discussed or places where it is genuinely rehearsed.

The same test will apply more broadly. Many organisations will discover that what they have been selling, rewarding or certifying was thinner than they thought. Others will use the shock to move closer to the parts of their work that are genuinely hard to imitate.

AI has not made management education obsolete. It has made some of its comfortable evasions harder to sustain. The issue is not whether business schools can keep up with a new technology. The issue is whether they can use this moment to become more honest about what they do, where their own value add lies and what they must now learn to do better. The answer to that question has become much harder to avoid.