There is a ghost haunting the server farms of Silicon Valley, though no one there would recognise it as a haunting. It is not the ghost of Marx, whose spectre the Valley’s libertarians long ago exorcised with stock options and disruption theology. It is more dangerous, and so profoundly misread that the misreading itself has become a civilisational event. It is the ghost of Friedrich Nietzsche—invoked constantly, understood almost never.
To grasp what has gone wrong, and why the wrongness of it matters now with an urgency unlike anything before, we have to return to the moment when human beings first decided that the world was something to be conquered.
The conquest that left the self intact
The Industrial Revolution was not, at its deepest level, a technical event. It was a philosophical one. When James Watt improved the steam engine in 1769, he inaugurated a new metaphysics: That nature exists to be mastered. Coal was torn from the earth’s ancient body; rivers were dammed and redirected; forests felled; mountains opened. The vocabulary that attached itself to this project was militaristic—industry conquered distance, subdued the wilderness, harnessed rivers. Nature was the adversary, and for the first time in human history, it appeared to be losing.
Marx saw what this cost the worker, the alienation of the labourer from the product of his labour, the reduction of human creativity to repetitive mechanical motion. But even within that brutalisation, something remained intact. The miner who descended into the earth was diminished by the descent; he was not dissolved by it. His consciousness, however narrowed and exhausted, remained his own. He went home. He dreamed. He felt hunger and desire and grief. The machine had captured his labour. It had not yet come for his interiority.
This distinction between the conquest of nature and the conquest of the self is the central distinction of our civilisational moment, and we have almost entirely failed to notice it.
Also read: Science once drove technology, now the reverse is true. Here’s how to harness it
The Übermensch they never read
Nietzsche published Thus Spoke Zarathustra between 1883 and 1885, in the full blaze of industrial modernity. He watched the smokestacks and railways proliferate across Europe, and what he saw disturbed him in ways the utopian progressivists of his age could not fathom. He did not object to transformation. What he objected to was the kind of human being that industrial civilisation was producing: the Last Man.
The Last Man is the anti-hero of Zarathustra’s opening speech, the creature who has invented happiness, who blinks, who seeks only comfort and the avoidance of suffering. He is not oppressed. He is worse than oppressed, he is satisfied. He has arranged his life so that nothing threatens him, nothing demands of him, nothing exceeds him. He has made himself, and this is Nietzsche’s most terrible word, manageable.
Against the Last Man, Nietzsche proposed the Übermensch, not as a century of misappropriation has insisted, a racial or biological category, not the blond beast of Nazi fantasy, but a philosophical ideal describing a mode of existence in which the human being takes full, creative, sovereign responsibility for its own values. The Übermensch does not receive meaning from God, from the state, from the market, from the machine. It creates meaning. And crucially, what it overcomes is itself, its own complacency, its own inherited values, its own comfortable smallness. The overcoming is self-directed. It is not technological.
When Elon Musk invokes enhancement, when transhumanists speak of uploading consciousness and merging with machine intelligence, they believe, with some fervour, that they are speaking Nietzsche’s language. They are speaking its precise opposite. They are describing the Last Man’s final achievement: The technological elimination of the necessity for self-overcoming at all.
The factory of the soul
The utilitarian tradition, which has dominated Anglo-American philosophy, and through its descendant, welfare economics, the entire architecture of policy and market rationality, begins from a different and entirely coherent premise: That the purpose of human existence is the maximisation of pleasure and the minimisation of pain. English philosopher Jeremy Bentham’s felicific calculus proposed that joy and suffering were quantities that could be added and subtracted with the same confidence as pounds and shillings.
It is no accident that this philosophical tradition emerged precisely when and where it did: In industrialising Britain, in the decades following the steam engine’s triumph. The utilitarian imagination and the industrial imagination share the same deep grammar. Both are optimising frameworks. Both treat complexity as a problem to be solved rather than a condition to be inhabited. Both are structurally hostile, not openly, but by their very architecture to the dimension of human experience that resists quantification: The tragic, the beautiful, the sacred, the meaningless.
By the time the effective altruism movement arrived at the end of the twentieth century, utilitarianism had become, in effect, a total theory of ethical management—identify the intervention that saves the most lives per dollar, measure, iterate, optimise. This is admirable in many particulars. It is also, in its deepest structure, the philosophy of the factory applied to the human soul.
Also read: Science lost something when it stopped asking ‘why’. It’s now high-tech animal husbandry
The colonisation of inwardness
The cyber revolution is routinely narrated as a second Industrial Revolution: Faster, cleaner, more democratic. Where the first overcame nature, this one overcomes distance and ignorance. Where the first created material abundance, this one creates informational abundance. The story is one of expansion, liberation, and acceleration.
This narrative is not false. It is incomplete in a way that is beginning to be catastrophic.
What the Industrial Revolution left intact, the human interiority that was subjected to but not dissolved by the machine, is precisely what the information revolution targets. Not deliberately, not maliciously, but structurally, by the logic of its own operation.
The attention economy, in its most honest description, is a set of technologies optimised to predict and manipulate human behaviour by modelling human psychology more accurately than human beings can model it themselves. The social media feed does not merely present information; it learns what you will respond to before you know that you will respond to it. It works on the pre-reflective, the emotional, the instinctual, the precisely human region that precedes the formation of conscious intention.
The philosopher Charles Taylor has argued that what defines the modern self is its inwardness, the sense that there is a space within us, a depth of interiority, that is genuinely our own, from which we form our values and make our choices. The Industrial Revolution impoverished that inwardness. The information revolution is attempting something more radical. It is modelling that inwardness, replicating it, and then using the model to colonise the original.
When a recommendation algorithm successfully predicts that you will click on a particular piece of content, when it knows, before you do, that you will be outraged by this headline, comforted by that image, seduced by this narrative, it has established a working model of your self that, in the relevant dimension, outperforms your own self-knowledge. At that point, the question of who is choosing, you or the model, becomes genuinely difficult to answer.
This is not the Übermensch. This is the Last Man at the moment of his perfection.
What the engineers cannot see
The enthusiasts of artificial general intelligence tend to reach for one of two frameworks when challenged to justify what they are building. The first is utilitarian. AI will cure diseases, accelerate scientific discovery, lift billions from poverty. The second is more grandiose. AI represents the next stage of evolution, the emergence of intelligence that transcends biological limitation. Both frameworks share a fundamental blindness. They conceive of intelligence as a resource, something that, like coal or oil, can be extracted, refined, and deployed in the service of predetermined ends. Neither asks the question Nietzsche considered the only serious one: What kind of human being does this technology produce?
The answer, emerging from three decades of evidence, is clarifying. The human being produced by the information environment is one whose attention is fragmented, whose memory is externalised, whose emotional responses are increasingly shaped by algorithmic amplification, and whose capacity for sustained, solitary, effortful thought the, capacity Nietzsche called the precondition for self-overcoming, is in measurable decline. The neuroscientist Maryanne Wolf has documented how deep reading, the kind that allows a reader to bring their own formed self into creative confrontation with a text, is being replaced by a shallower scanning mode adapted to the rhythms of the screen. This is not a metaphor. It is a neurological observation.
What is being lost is not information. We have never had more information. What is being lost is the capacity to do anything with information that requires a sustained, reflective, and internally coherent self. What is being lost, in Nietzsche’s precise vocabulary, is the precondition for overcoming.
Also read: India has a science soft power gap. Even after giving the world zero, Covaxin, Chandrayaan
The myth of augmentation
There is a word that has entered the culture recently, deployed mostly with enthusiasm of frivolity: Augmentation. The AI does not replace the human, the argument goes; it augments them. The radiologist with the AI diagnostic assistant sees more, misses less, saves more lives. The student with the AI tutor learns more efficiently, masters more material.
All of this may be true in the narrow sense, in which the industrial analogy was also true: The factory worker with the power loom produced more cloth per hour than the hand-weaver. Productivity increased; material welfare improved. And yet something was lost, something the Luddites, so routinely mocked by the historians of progress, actually understood. The hand-weaver’s knowledge, the bodily intelligence embedded in the craft, the web of social and economic relations organised around the practice of weaving, constituted a form of human life whose value was not captured by the quantity of cloth produced.
The augmentation argument for AI makes the same error at a deeper level. When the student uses an AI to master calculus more efficiently, what is being optimised is the output, performance on assessments, ability to apply techniques. What is not being measured, and what may in fact be diminished, is the cognitive struggle through which mathematical understanding, as opposed to mathematical competence, is formed. The confusion, the false starts, the moments of breakthrough, the gradual formation of an intuition that becomes genuinely one’s own, this is not a bug in the educational process. It is, in large part, the process. It is what Nietzsche had in mind when he wrote that one must have chaos within oneself to give birth to a dancing star.
The efficient delivery of correct answers to a student who has not endured the chaos does not produce a mathematician. It produces, at best, a competent calculator—and, at worst, someone who cannot tell the difference.
The circle Nietzsche saw
The philosopher Bernard Stiegler argued that the relationship between human beings and technical systems has always been constitutive of human identity, that we have never been human in the absence of tools. This is true. The book, after all, is a technology, and the formation of the literate self through the encounter with books is no less mediated for being centuries old.
But Stiegler also argued, and this is the part of his thought least absorbed, that not all technical systems are equivalent in their effects on human interiority. Some technologies support the processes of self-formation. Others short-circuit them. The book invites the reader to bring their formed self into productive collision with another formed self. The algorithmic feed invites the user to dissolve into a flow state calibrated to maximise engagement. These are not the same relationship. The difference is not quantitative but qualitative: One supports the formation of a self capable of genuine encounter; the other systematically erodes the conditions for that formation.
This is the argument the utilitarian framework cannot make, because it cannot register qualitative differences of this kind. The utilitarian can count hours of reading against hours of scrolling, measure self-reported wellbeing, and conclude that if people prefer scrolling, scrolling produces more utility. What the utilitarian cannot do is ask whether the preference itself has been manufactured by the technology, and whether a self whose preferences have been algorithmically moulded is the kind of self whose preferences it makes any sense to satisfy.
This is the circle that Nietzsche saw, a hundred and forty years before the algorithm. ‘What is great in man,’ Zarathustra says, ‘is that he is a bridge and not an end.’ The Übermensch is not a destination; it is a direction, the direction of perpetual self-transcendence, of the self that overcomes its own inherited, conditioned, and comfortable smallness. The information age, at its current trajectory, is not building that bridge. It is offering a very comfortable place to sit down.
The factory returns
None of this is an argument for primitivism, or for some fantasy of technological uncoupling. The conquest of nature the Industrial Revolution accomplished was genuine liberation for billions of people. The informational abundance of the cyber revolution contains real goods, access to knowledge, to art, to community, to scientific discovery, that previous generations could not have imagined.
The argument is against a specific philosophical error, the belief that what the technology is doing is enhancing the human, when what it is doing, at the structural level, by the logic of its own incentives, is constructing a more manageable version of the human. A creature whose preferences are known, whose behaviour is predictable, whose attention is capturable, and whose interiority has been modelled with sufficient accuracy that the distinction between the genuine self and the simulation of the self has become, for practical purposes, irrelevant.
This is not overcoming. It is epistemic eradication, the systematic dissolution of the conditions under which genuine knowing, genuine willing, genuine self-formation are possible. And it is being built, at scale, with our enthusiastic consent.
Zarathustra descends from his mountain full of warnings that no one wants to hear. He speaks of the Last Man and is met with applause: The crowd thinks he is describing an ideal, not an abyss. There is something very contemporary about this scene, the enthusiasts who mistake the warning for a promise, who look at the description of what we are becoming and say: Yes, exactly, that is what we are building.
But Zarathustra does not stay to argue with the crowd. He withdraws, because he understands that the self cannot be saved by argument. It can only be saved by the practice of those who refuse to be managed, who insist, against every incentive, on the friction, the confusion, the sustained and costly encounter with difficulty that alone produces a self worth the name. The Übermensch is not a posthuman destination. It is the decision, made again every day, to remain genuinely human: unmodelled, unoptimised, and unfinished.
The factory has returned. It has come, this time, not for the labour of the body but for the formation of the mind. What we owe the future is not a better algorithm. It is the refusal to be its raw material.
Pranav Sharma is a historian of science who lives and writes from New Delhi, India and Paro, Bhutan.
(Edited by Theres Sudeep)