There is a particular kind of anxiety that now accompanies education in the age of artificial intelligence (AI). It is not merely the old anxiety about examinations, competition or jobs. Those fears belonged to a familiar world in which education, however unequal, was still understood as the acquisition of capacities that machines did not possess. A student learned to write, calculate, compose, analyse, design or argue, and these abilities were valued because they remained recognisably human.
That assumption is now difficult to sustain. Saikat Majumdar’s Open Intelligence: Education Between Art and Artificial begins from this unsettled moment. The book is neither a celebration of AI nor a lament for a vanishing human world. It asks what education might mean when machines enter territories long regarded as the preserve of human intelligence: writing, imagination, analysis, composition and creative experimentation.
The most important question Majumdar raises is therefore not whether AI will become more capable. It almost certainly will. The harder question is what human beings should continue to learn when competence itself is no longer a sufficient guarantee of distinction.
One of Majumdar’s most illuminating arguments concerns artistic and creative labour. Every art has a technical dimension. There are rules of syntax and grammar, conventions of narrative, structures of rhythm and metre, principles of composition and countless techniques that can be taught, practised and improved. But art is not exhausted by technique.
There is another dimension that emerges from individual experience, temperament, memory, perception and circumstance. It is what makes two people, given the same technical resources, produce radically different works. It is also what makes a work recognisable as belonging to a particular consciousness.
Generative AI has complicated this distinction by showing how much of what we once considered creative competence can be modelled and reproduced. A language model can write an essay, poem, screenplay or business document in seconds. The results may not always be profound, but they are often competent enough to enter domains once protected by the assumption that creativity required exclusively human intelligence.
This creates a crisis for education. Classrooms have long concentrated on technique: writing an argument, structuring an essay, solving a problem and producing the expected answer. Yet these are precisely the areas in which machines are becoming efficient. Technique remains necessary, but it can no longer be education’s final destination.
Majumdar’s defence of human creativity does not depend on claiming that machines cannot produce anything resembling art. AI systems already generate images, music and prose that may surprise experienced practitioners. The more interesting question is the difference between producing an expression and having a life from which that expression emerges.
A machine can generate a description of grief. It can reproduce the vocabulary of mourning, loss or separation and identify the narrative structures through which human beings have represented such experiences. What remains unsettled is whether this generation of language can ever amount to the experience itself.
That distinction matters because art has always been connected to the particular. A work acquires force not simply because it conforms to a pattern but because something singular has passed through it: a memory, an encounter, a place, a historical wound, a private obsession, a way of seeing. This does not mean human beings possess some mystical territory machines can never approach. It means creativity cannot be reduced to output. We must also ask about experience, intention, context and consequence.
The human signature may therefore lie less in the ability to produce something a machine cannot produce and more in the reasons for which we choose to produce it. That is a far more demanding proposition.
The debate about AI is often framed as a contest between humans and machines. Majumdar’s argument becomes more compelling when placed within the larger economic system in which the technology is deployed. AI can transform work across publishing, journalism, education, law, advertising, design, finance, medicine and other fields built around information.
But technology alone does not determine the future. Markets do. Contemporary institutions often reward efficiency more readily than originality. Publishers need books that can find readers. Film studios need stories that can attract audiences. Corporations seek predictable returns. Educational institutions increasingly speak the language of employability, rankings and measurable outcomes.
In such a system, AI possesses an obvious advantage. It can produce enormous quantities of competent material quickly and at relatively low cost. The danger is not simply that machines will replace human artists or professionals. It is that institutions may begin to prefer machine-generated adequacy to difficult human originality.
A culture does not lose its imagination in one dramatic event. It may lose it gradually through rational decisions: a publisher choosing the safer manuscript, a studio repeating a successful formula, a classroom rewarding the expected answer. The machine then becomes not the cause of cultural standardisation but its most efficient instrument. The greater danger is a degradation of taste, where fluent and predictable content becomes enough.
This is where Open Intelligence acquires its greatest significance. Education has always carried two contradictory possibilities. It can liberate the mind, but it can also discipline it into conformity. It can encourage curiosity, or it can turn learning into the accumulation of credentials. It can cultivate judgement, or it can reward the ability to produce the expected answer.
The Indian education system knows this contradiction well. For millions of students, education remains tied to examinations, rankings, degrees and employability. In an unequal society, standardised examinations can offer the appearance of a common measure and a route into institutions that might otherwise remain inaccessible.
But what happens when the abilities most easily measured by such systems are also the abilities AI can increasingly perform? If a system can write an essay, analyse a document or produce a presentation, education cannot merely ask whether a student can complete the task. It must ask whether the student understands why the task matters, whether the result is trustworthy and what assumptions lie beneath it.
This is where Majumdar’s idea of Open Intelligence becomes important. The phrase suggests an intelligence not closed within fixed categories of knowledge or predetermined measures of ability. It points towards an education capable of accommodating curiosity, experimentation, contradiction and forms of thought that cannot easily be converted into a standardised metric. Such an education would not teach students to compete with machines at being machines. It would teach them to recognise what questions are worth asking.
Educational institutions often treat AI mainly as a problem of academic integrity: students use ChatGPT, institutions respond with restrictions and detection software. But this misses the larger transformation. If students will live and work alongside systems that generate language, images, code and analysis, banning them cannot be an educational philosophy. Nor can treating AI as an oracle.
The real task is to develop judgement. Students will need to formulate questions, examine assumptions, recognise bias, verify information, understand the limits of generated content and decide when not to use a machine at all. They will need to become editors, interpreters and ethical decision-makers rather than passive consumers of machine-produced answers.
This also changes the meaning of expertise. The expert of the future may not be the person who can produce the greatest volume of information, but the person who can distinguish the significant from the trivial, the reliable from the plausible, the original from the derivative. That is not a diminished conception of intelligence. It is a more difficult one.
Beneath the debate over skills lies a larger issue. Modern societies increasingly connect human worth with productivity: what someone does, earns and produces. AI exposes the limits of that arrangement. If machines can perform more economically valuable tasks, what happens to a society that defines human worth almost entirely through productivity?
Majumdar’s answer is not to retreat from technology but to recover a larger understanding of education. Education must create people capable not merely of earning a living but of deciding what constitutes a life worth living. It must cultivate attention, imagination, judgement and ethical responsibility. This is not an argument against work, but against making work the sole measure of a person.
The phrase Open Intelligence ultimately acquires its force because it describes a possibility rather than a finished doctrine. Intelligence, in this conception, cannot be reduced to solving predefined problems or producing correct answers. It must include the capacity to encounter uncertainty, move between disciplines, recognise one’s limitations and remain open to forms of knowledge that do not fit comfortably within existing categories.
This is especially important in the age of generative AI because machines are changing not merely what we can do but the hierarchy of skills through which education has traditionally been organised. The challenge before universities and schools is not simply to add an AI course to an existing curriculum. It is to reconsider what the curriculum is for.
Perhaps students should spend less time learning how to produce the first acceptable answer and more time learning how to recognise a better question. Perhaps the classroom should become a place where uncertainty is not treated as failure. Perhaps education should protect precisely those dimensions of human experience that resist immediate measurement.
These are not easy reforms. They require institutions to surrender some attachment to predictability and measurable outcomes. They also require educators to acknowledge that they are entering unfamiliar territory. That may be the most valuable lesson of Majumdar’s book.
Open Intelligence is ultimately a humane book because it refuses to reduce the future to a contest between human beings and machines. The question is not whether AI will become more intelligent. The question is what we will do with the intelligence that machines increasingly make available to us.
AI may deepen the worst tendencies of contemporary culture: speed over reflection, scale over particularity, efficiency over judgement and marketability over imagination. Yet it may also free human beings to attend to questions machines cannot settle: what is meaningful, what is just, what is beautiful and what kind of society we wish to inhabit.
The choice will not be made by technology alone. It will be made in classrooms, publishing houses, universities, corporations and cultural institutions. It will be made by teachers deciding what deserves attention, by editors deciding what deserves publication, by employers deciding what they value and by societies deciding what they mean when they speak of talent.
Majumdar’s important contribution is to move the conversation away from the familiar question of whether AI can imitate human creativity. That question may become less interesting as the technology improves. The more consequential question is whether human beings can preserve the conditions in which creativity, judgement and curiosity remain valuable.
The future of education may therefore depend not on teaching students to outrun the machine, but on helping them understand why they are running at all.
Book: Open Intelligence: Education Between Art and Artificial
Author: Saikat Majumdar
Publisher: Penguin Random House India
Price: INR 399/