When Intelligence Becomes a Commodity

What happens to the university when the thing it was built to distribute—intellectual capability—becomes abundant?

For generations, higher education rested on a powerful economic bargain. Knowledge was difficult to obtain, expert instruction was scarce, and educated intellectual labour was valuable. A student could therefore make a four-year investment in education with a reasonable expectation that the degree would open a predictable path into work, income and status.

Artificial intelligence is beginning to disturb every part of that bargain. The important question is not simply whether AI will replace particular jobs. It is whether the economic value of the human intellectual labour that universities have traditionally prepared people to sell will remain scarce.

1. The Pipeline That Made Education Valuable

In India, the traditional pipeline is unusually visible: JEE, IIT, degree, campus placement, career, income and status. The university sits in the middle of this pipeline as a trusted intermediary between knowledge and economic opportunity. It teaches a body of knowledge, certifies competence and gives employers a relatively reliable signal about the person carrying the degree.

That system worked because the underlying resource was scarce. A student could not simply summon a professor, a laboratory, a research assistant, a software engineer or a mathematical expert on demand. Universities aggregated these scarce resources and made them accessible through a structured institution.

AI changes the economics of that arrangement. A capable student can increasingly obtain explanations, examples, code, analysis, research assistance, writing support and personalised tutoring at almost any hour. The university therefore faces a fundamental problem: if answers and much of the cognitive labour behind those answers become cheap, what exactly is the student paying the university to provide?

The danger is not that IIT becomes useless. The danger is that IIT may become something very different from what students think they are buying.

2. The Soft Brain Arrives

The first stage is not replacement. It is augmentation. AI behaves like a soft external brain: it helps a person search, calculate, write, code, compare, explain, translate and explore. The existing human system becomes faster and more capable without changing its basic architecture.

This matters because universities have traditionally trained people to perform many of these cognitive tasks themselves. If AI can perform a large part of the task reliably, the economic value shifts upward. The question becomes less about whether a person can execute the task and more about whether the person can define the task, supervise the system, evaluate the result and decide what should happen next.

This is why the AI transition should not be described only as job replacement. It is also a movement of human contribution up the stack.

3. When the Four-Year Bet Becomes Uncertain

The traditional university promise contains an implicit prediction: what you learn today will remain sufficiently valuable when you graduate. AI makes that prediction harder to defend. A fourth-year student can reasonably ask whether the job they expected will still exist at graduation. A second-year student can ask whether the jobs they are preparing for will survive another two years. A first-year student can ask the more uncomfortable question: am I spending four years preparing for a labour market that may be structurally different by the time I arrive?

This is why anxiety can travel backwards through the entire student pipeline. The uncertainty is not confined to graduates. It reaches the decision to enter the pipeline itself.

The IIT paradox follows. AI may make an exceptional student far more powerful, while simultaneously reducing the need for large numbers of conventional intellectual workers. The institution could become more intellectually valuable even as its traditional credential-based employment value becomes less predictable.

4. From Fixed Curriculum to Adaptive Umbrella Curriculum

A fixed curriculum assumes that an institution can decide, several years in advance, what knowledge a student will need later. That assumption becomes increasingly fragile when AI and technology are moving continuously.

The alternative is an adaptive umbrella curriculum. The umbrella remains stable: engineering, medicine, economics, law, physics, design or another broad domain. What moves underneath it is the frontier—tools, applications, research directions, methods, industry practices and emerging problems.

AI can help make this practical. It can monitor developments, identify gaps, personalise learning paths, generate exercises, test understanding and suggest what a particular student should explore next. The institution provides the stable intellectual framework; the curriculum becomes a living system rather than a four-year package frozen at admission.

The umbrella stays; the curriculum moves.

5. From Knowledge Provider to Human Network

If knowledge becomes abundant, the university's scarce resources become easier to see. Exceptional peers, laboratories, physical research environments, certain mentors, reputation and networks are scarce resources. They cannot be copied simply by opening another browser tab.

Consider the old metaphor of knowledge as water. The university used to control an important part of the pipeline to the reservoir. AI is changing that. Everyone increasingly has a pipeline to the sea.

Who will buy water now?

The university can no longer justify itself primarily by controlling access to information. It has to explain why its particular shore is worth visiting. That may mean access to exceptional people, difficult experiments, high-trust collaboration, ambitious projects, institutional reputation and opportunities that are genuinely difficult to reproduce alone.

This also changes the meaning of campus. The physical institution may become less important as a place where lectures are delivered and more important as a place where people, experiments, projects and opportunities collide.

6. The Mind Moves Up the Stack

Software development provides a useful illustration. The progression can be imagined as Software Developer, AI-Assisted Developer, Prompt Engineer, Idea Prompt Developer and eventually Idea Selector.

At each stage, the scarce human contribution moves upward. First the person writes code. Then the person directs AI to write code. Then the person learns to express an intention precisely enough for AI to execute it. Then the person begins prompting at the level of ideas. Eventually the central task may be deciding which ideas deserve attention at all.

That final shift matters because AI can attack two traditional human weaknesses at the same time: information scarcity and information fatigue. It can search areas we did not think to search, challenge assumptions, expose blind spots and bring ideas from distant disciplines. But abundance creates its own problem. If AI can produce a thousand plausible ideas, the bottleneck becomes selection.

The human role therefore moves toward judgment: what matters, what is true enough to act on, what is worth building, what should be rejected and what objective should govern the system in the first place.

7. AI as Reductio ad Absurdum

There is a broader pattern here. AI does not merely automate an existing machine. It can push that machine toward its logical endpoint.

The sequence is simple: Tighten. Replace. Revamp.

First, AI tightens the existing system. It finds waste, duplication, bottlenecks and unnecessary steps. Next, it replaces individual components with cheaper or better components. Finally, once enough components have changed, the architecture itself becomes questionable.

Why keep the machine in its original form at all?

The aviation analogy is useful. An aircraft may begin with two pilots because that is the established architecture. Automation first assists them. Then it can reduce workload. Eventually the question becomes whether two pilots are technically necessary at all, or whether one human, remote oversight or autonomous control could perform the function with greater reliability.

The same logic can reach law. AI can improve legal research and argument, but the deeper transformation may be earlier in the chain. Before a dispute reaches court, an AI system could simulate both sides, estimate the probability of winning, model likely damages and costs, identify evidence that could change the outcome and propose a settlement range. The technological opportunity is not merely to optimise the courtroom. It is to make fewer people need the courtroom.

This is what makes AI a kind of technological reductio ad absurdum. It takes systems built around scarcity, friction and human cognitive limits and asks, implicitly: if these constraints disappear, why does the system still need to look like this?

8. What Is the University For?

The answer cannot simply be 'to provide knowledge.' Knowledge is becoming too abundant for that to remain a sufficient explanation. Nor can the answer be 'to train people for today's jobs,' because today's jobs may not be stable enough to justify a four-year forecast.

The future university must help people navigate abundance. It should develop the ability to formulate important problems, work with AI, test ideas, conduct experiments, exercise judgment, collaborate with exceptional people and recognise opportunities that are not yet obvious.

That changes assessment too. If AI can produce an excellent essay or solve a standard problem, reproducing the answer is no longer strong evidence of capability. Institutions will increasingly need to evaluate understanding through application, oral defence, experimentation, project work, problem formulation and the ability to explain why a particular answer or direction was chosen.

The university therefore moves from being primarily a knowledge-distribution system to being an environment for capability formation.

9. Conclusion: The University After Intelligence Scarcity

Universities were built for a world in which knowledge was difficult to access and intellectual labour was expensive. AI is changing both conditions. That does not make education irrelevant. It makes the old justification incomplete.

An institution that simply defends its inherited model will increasingly look like a system designed for a previous economy. An institution that adapts can become more valuable, not less: a place where people learn to operate with powerful machines, work on problems that matter, gain access to difficult physical and social environments, and develop judgment in circumstances where the answer is not already known.

The university may survive AI. But it cannot assume that the university we inherited is the university we will need.

10. Guidance for the Future

For students, the sensible strategy is not to predict the exact job that will exist four years from now. Build strong fundamentals, learn to work with AI rather than around it, practise selecting and framing problems, and develop the ability to move when the frontier moves.

For universities, the priority is structural rather than cosmetic. An adaptive umbrella curriculum, AI-native learning, flexible assessment, stronger research exposure and more deliberate access to exceptional people and environments are likely to matter more than simply adding an AI elective to an old degree.

For educators, the value shifts from being the person who possesses information to being the person who can challenge a student, expose a blind spot, improve a question, supervise difficult work and help turn curiosity into capability.

For employers, the degree signal will increasingly compete with evidence of what a person can actually accomplish with AI. The important question becomes less 'What did this person study?' and more 'What can this person build, discover, decide or change?'

And for a student choosing a university, perhaps the most useful question is no longer simply 'Which college has the best placement record?' It is: 'What will I be able to become here that I could not become as effectively on my own with AI?'

Once substantial intellectual capability becomes abundant, the scarce resource is no longer access to answers. It is knowing which questions are worth asking—and having the people, environment and courage to pursue them.

Perspectives behind the question

This article was prompted in part by a candid conversation between Prof. Prathosh A.P. and his students at IISC Bangalore in Sep 2026, about the purpose of higher education in the age of AI.

You can watch the discussion.

MIT students have been asking related questions from inside the university itself; an April 2026 MIT panel brought students together to discuss how AI is changing learning, its impact on their education, and the ethical questions surrounding its use.

Read the MIT student discussion.

Research is changing too

Research is not outside this transformation. AI tools are beginning to change how new knowledge is produced: MIT reports that researchers are already using AI to generate promising hypotheses and pressure-test possible solutions, potentially accelerating discovery.

The university therefore has to rethink not only how it teaches existing knowledge, but also how it supports the production of new knowledge. That strengthens the case for experimentation, problem formulation, evidence evaluation and human judgment as core university capabilities.

MIT's 2026 statement on AI in research and MIT research on technology, new work and expertise are useful reference points.