Even the best weapon
is an unhappy tool,
hateful to living things.

Tzu & Le Guin, Tao Te Ching, page 38

“Artificial intelligence” is not a technology. A chef’s knife is a technology, as are the practices around its use in the kitchen. A tank is a technology, as are the ways a tank is deployed in war. Both can kill, but one cannot meaningfully talk about a technology that encompasses both Sherman and santoku; the affordances, practices, and intentions are far too different to be brought into useful conversation. Likewise, in the hysterical gold rush to hoover up whatever money they can, the technocrats have labeled any and all manner of engineering practices as “AI” and riddled their products with sparkle emojis, to the extent that what we mean when we say AI is, from a technology standpoint, no longer meaningful. AI seems to be, at every moment, everything from an algorithm of the kind that has been in use for half a century, to bullshit generators that clutter up our information systems, to the promised arrival of a new consciousness—a prophesied god who will either savage us or save us or, somehow, both at the same time. There exists no coherent notion of what AI is or could be, and no meaningful effort to coalesce around a set of practices, because to do so would be to reduce the opportunity for grift.

What AI is is an ideology—a system of ideas that has swept up not only the tech industry but huge parts of government on both sides of the aisle, a supermajority of everyone with assets in the millions and up, and a seemingly growing sector of the journalism class. The ideology itself is nothing new—it is the age-old system of supremacy, granting care and comfort to some while relegating others to servitude and penury—but the wrappings have been updated for the late capital, late digital age, a gaudy new cloak for today’s would-be emperors. Engaging with AI as a technology is to play the fool—it’s to observe the reflective surface of the thing without taking note of the way it sends roots deep down into the ground, breaking up bedrock, poisoning the soil, reaching far and wide to capture, uproot, strangle, and steal everything within its reach. It’s to stand aboveground and pontificate about the marvels of this bright new magic, to be dazzled by all its flickering, glittering glory, its smooth mirages and six-fingered messiahs, its apparent obsequiousness in response to all your commands, right up until the point when a sinkhole opens up and swallows you whole.

The “artificial” in AI is a MacGuffin. The taproot of this ideology is intelligence: that is, an intelligence that can be measured and ranked, an intelligence that is both quantifiable and presumptively quantified. Presumptively, because those asserting its value rarely bother with the measurement itself, preferring instead to infer it from other characteristics. Being in a position of authority, if one is wealthy, cisgendered, male, and white, is taken as a priori evidence of brilliance. If one is none of those things, then any status is presumed to be unauthorized, a trick, a manipulation, a witch disguising herself with a glamour—the dreaded “DEI-hire.”

This notion of a hierarchical intelligence appears first as tautology, second as farce. European colonists arrived on rich lands filled with people who were not like them, and conveniently concluded both that the land was for the taking and that the people were stupid and savage, barely human or perhaps not even human at all. In Superior: The Return of Race Science, Angela Saini notes that,

Kant stated in Observations on the Feeling of the Beautiful and Sublime in 1764, “The Negroes [sic] of Africa have by nature no feeling that rises above the trifling.” When he met a quick-witted carpenter, he quickly dismissed him with the observation that, “this fellow was quite black from head to foot, a clear proof that what he said was stupid.”

Saini, Superior, page 10

Kant was, of course, hardly alone in his vile beliefs. But unadorned racism was not then, as it is not now, always without reproof. As such, its proponents sought justification, in the form of a sheen of scientific evidence with which they could enrobe their despicable claims. In the nineteenth century, with the abolition movement ascending, racist scientists took to measuring brain size—first with calipers on the heads of living subjects, then by filling skulls with mustard seed and lead shot, and later by weighing the brains of deceased scientists who donated their bodies to the cause. That brain size should correlate with increased intelligence was never more than supposition, and one assumes the practitioners of craniometry didn’t consider the sperm whale their better. Regardless, the data refused to corroborate their claims without the aid of manipulation and statistical errors, and the weighing of brains proved an embarrassment when some of the great men of the day were discovered to be lightweights.¹

If the size of brains couldn’t be adequately measured and ranked, then surely the function of them could be, followed the unimpeachable thinking. Tossing aside calipers and scales, a new generation reached instead for interrogation. As Stephen Jay Gould catalogues in The Mismeasure of Man, the first intelligence tests were created by Alfred Binet in 1905, and designed to ascertain if a school child was falling behind their peers so that they could be afforded additional tutoring. Binet went to lengths to assert that his test didn’t measure something immutable, nor that it should be used for anything other than the intended narrow purpose; but the power of a belief in the biological determination of intelligence was no match for his disclaimers. In 1916, L. M. Terman adapted Binet’s exam into the Stanford-Binet IQ test—the progenitor of every IQ test since—and asserted its suitability for identifying and dealing with the problem of the “feeble-minded.” Unlike Binet, Terman was an avowed eugenicist, convinced that intelligence was not only hierarchical but that the hierarchies were embedded in blood and skin; moreover, he believed that people of lesser intelligence should necessarily be treated as unequal to their wiser brothers. As he put it,

It is safe to predict that in the near future, intelligence tests will bring tens of thousands of these high-grade defectives under the surveillance and protection of society. This will ultimately result in curtailing the reproduction of feeble-mindedness and in the elimination of an enormous amount of crime, pauperism, and industrial inefficiency.

Gould, The Mismeasure of Man, page 209

A year later, a Harvard psychologist named Robert Yerkes would tap both Terman and Carl Brigham of Princeton—like many of his brethren, a staunch believer in the need for immigration controls to preserve American’s intellectual capacity.² Yerkes convinced the Army that they should test the IQ of every Army recruit. Nearly two million soldiers were evaluated during the subsequent years, but the testing conditions were beyond absurd: written tests were administered to illiterate recruits or those who had never held a pencil; oral tests included rapidly spoken and bewildering instructions in English—a language that many recruits did not natively speak. The test purported to analyze innate intelligence, not cultural learning, but the questions asked after such subjects as American baseball players, the design of the phonograph, and Crisco.³ Yerkes and his collaborators subsequently manipulated the numbers in order to conclude that the test results gave proof to the validity of a racial hierarchy of intelligence, with white Americans scoring higher than darker-skinned European immigrants, and Black Americans scoring at the bottom. This despite the fact that the data included abundant prima facie evidence of the test’s failure, including that the most common total score across numerous groups was zero.⁴

Brigham would go on to develop the SAT exam in order to assess students’ fitness for college, using many of the same principles (and reproducing the same biases) as were deployed in the Army test. He would ultimately recant his involvement in the Army test, issuing a rare apology, and acknowledging that the “study with its entire hypothetical superstructure of racial differences collapses completely.”⁵ But other pursuers of rankable intelligence would never accept the obvious flaws in their thinking: Cyril Burt, then Chair of Psychology at University College, London, went so far as to fabricate data to support his hereditarian theory of intelligence. Burt’s fraud wasn’t uncovered until after his death, many years after others had repeatedly used his results to assert the “inherited and ineradicable differences in intelligence between whites and blacks [sic] in America.”⁶ Over and over in the pursuit of quantified intelligence we see errors of logic, miscalculations, manipulations of data, and overt and outright fraud—alongside practitioners who refuse to reject the faith. As with every proper doomsday cult, proof is always deferred.

It’s this belief in a measurable and hierarchical intelligence that leads directly to declarations like this one, from Sam Altman, in February of this year:

The socioeconomic value of linearly increasing intelligence is super-exponential in nature. A consequence of this is that we see no reason for exponentially increasing investment to stop in the near future.

This is an astonishing statement, not least for its complete lack of self-awareness, an example of what Chris Xu aptly refers to as blowhard syndrome. Note the two moves here: first, the assumption that intelligence is a measurable quality in the first place, and, following from that, that increasing intelligence necessarily justifies (exponentially!) increasing resources.

This was always the point! More supposed smarts equals more money, more care, more life. The flawed tests and fake data were never more than a thinly-disguised means to the violent enforcement of inequality. In their stead we now have blandly grandiose assertions delivered by technocrats whose own intelligence is likewise assumed. But the end is the same: super intelligence demands super concentrations of capital, the same captains of industry deepening their planet-sized pockets while simultaneously issuing layoff after layoff, resegregating workplaces, and threatening the livelihood of workers everywhere.

It’s instructive that one of the mechanisms for perpetuating this ideology are chattering bots that speak both fact and falsehood in the same servile and confident tone, their makers unconcerned with the difference. In fact, their makers seem entirely concerned with obviating that difference, with disappearing distinctions between knowledge and ignorance, without which truth becomes entirely a product of power. Proving the superiority of some humans over others has repeatedly failed; what better way to continue the effort than the deployment of technology that makes proof of anything impossible, such that making something true requires only the right person to declare it so.

We see such declarations in repeated assertions of Elon Musk’s high IQ, despite no test results ever being shared (or indeed, even any evidence that he has ever taken the test, or that he took it without cheating); in Trump’s insipid declarations that every strongman he meets is “very smart, very handsome”; in his pervasive denigration of Black women, especially those who have, in his eyes, unjustly risen to positions of authority. We see it also in the evisceration of due process as people of color are violently rendered to countries with majorities of the same, alongside the performative welcoming of white refugees amidst an entirely falsified narrative of “white genocide” (a narrative echoed by the bots themselves). We see it in AI’s present-day—not hypothetical—use to deny medical care to people in need, against the judgment of their human doctors and caretakers. We see it in accelerated investments in fossil fuels, a choice completely at odds with what we know of climate change and our available means to slow it down and preserve a habitable planet for all.

Intelligence has never been an objective quality that can be ascertained the way we measure the (actually increasing) carbon in the atmosphere. It is a political device that preserves power and care for those deemed worthy of it, and which simultaneously withdraws such care from everyone else. Its latest incarnation, with that modifier artificial, asserts its power through programs that wash accountability from the programmers: control, wealth, and power run up to founders and investors while harms run down to the rest of us with no possibility of redress. The ideology beneath AI is nothing more than white supremacy glammed up by a techbro aesthetic, white hoods swapped for tactical pants, the old racial slurs upgraded to anti-DEI rhetoric, the burning cross traded in for midnight reductions in force.

At midnight on Friday March 1, every remaining member of 18F was abruptly laid off, and the eleven-year-old digital agency was shuttered. 18F might seem like an odd target for Elon Musk’s DOGE: the tech collective had a demonstrated track record of saving the government money, through efficient application of technology and associated practices. Any actual interest in government efficiency would have naturally led to steep investments in 18F, or efforts to reproduce the model across state and local governments.

But efficiency, like intelligence, was only ever a cover. The decimation of 18F makes a lot more sense when you recognize the threat that 18F posed to the ideological efforts underpinning DOGE and the rest of the “AI-first” movement: here was a group of talented technology practitioners with deep expertise across both the government and corporate sectors, who dedicated their time and skill to making government services accessible to all, especially those most in need. That is, they enabled care and support for those the modern-day eugenicists deemed worthy only of control, denigration, and impoverishment. It’s no contradiction that the same DOGE minions tasked with increasing “efficiency” immediately set upon removing pronouns from email signatures and deleting websites that spoke to gender and racial equality. The threat that 18F posed wasn’t to efficiency, it was to inequality. And it’s inequality that AI propagates, just as it was inequality that the skull measurements and IQ tests perpetuated.

AI hype men frequently exhort that AI is coming for all our jobs, and only those clever enough to learn how to deploy it will survive. That this is a plainly self-serving statement is enough to dismiss it out of hand, but even beyond that, there remains little to no evidence that it’s anything but a pipe dream, as the regular drumbeat of backtracks attests. The more likely scenario, and the one that’s been exhibited time and again with automation technology in previous centuries, is that the work that remains will be deskilled, de-spirited, stripped of creativity and joy, and granted the meanest remuneration.

L. M. Terman, whose aim in creating the Stanford-Binet IQ test was to weed out undesirables, remarked that:

The evolution of modern industrial organization together with the mechanization of processes by machinery is making possible the larger and larger utilization of inferior mentality. One man with ability to think and plan guides the labor of ten or twenty laborers, who do what they are told to do and have little need for resourcefulness or initiative.

Gould, The Mismeasure of Man, page 212

This has always been the intention of AI, and where its connection to the intelligence-rankers of years past is cruelly apparent: if those in power cannot prove that a great many people are already inferior then they will bring that inferiority about by forcing them to use a tool that diminishes their intellectual and creative capacity. I think of the engineers and designers who have spent decades honing their skills, deepening personal and public creative practices in service both to the users of the systems they built and to their own brilliant spirits, now being told to park themselves in front of a sycophantic oracle that can be appeased only through rote dictates, and which never tires of lying even as their own minds and muscles atrophy from disuse. What is being automated here: the work or the people?

In the third book in Ursula K. Le Guin’s Hainish sequence, City of Illusions, Earth has been taken over by a people known as the Shing. Across the planet, humans live in small villages or else move nomadically with their herds, prevented from knowing their history or from rebuilding the ancient cities and knowledge that the Shing razed and erased. Meanwhile the Shing, themselves very small in number, live sheltered in an extraordinary skyward city, perched above a canyon in buildings made of semi-transparent material. When Ramarren, a visitor from a distant planet, arrives on earth, he awakens in this terrible place:

He was alone, in the midst of a room so uncanny that it revived his dizziness for a while. There was no furniture. Walls, floor and ceiling were all of the same translucent stuff, which appeared soft and undulant like many thicknesses of pale green veiling, but was tough and slick to the touch. Queer carvings and crimpings and ridges forming ornate patterns all over the floor were, to the exploring hand, nonexistent; they were eye-deceiving paintings, or lay beneath a smooth transparent surface. The angles where walls met were thrown out of true by optical-illusion devices of cross-hatching and pseudo-parallels used as decoration; to pull the corners into right angles took an effort of will, which was perhaps an effort of self-deception, since they might, after all, not be right angles. But none of this teasing subtlety of decoration so disoriented [him] as the fact that the entire room was translucent. Vaguely, with the effect of looking into a depth of very green pond-water, underneath him another room was visible. Overhead was a patch of light that might be the moon, blurred and greened by one or more intervening ceilings. Through one wall of the room strings and patches of brightness were fairly distinct, and he could make out the motion of the lights of helicopters or aircars. Through the other three walls these outdoor lights were much dimmer, blurred by the veilings of further walls, corridors, rooms. Shapes moved in these other rooms. He could see them but there was no identifying them: features, dress, color, size was all blurred away. A blot of shadow somewhere in the green depths suddenly rose and grew less, greener, dimmer, fading into the maze of vagueness. Visibility without discrimination, solitude without privacy. It was extraordinarily beautiful, this masked shimmer of lights and shapes through inchoate planes of green, and extraordinarily disturbing.

Le Guin, Worlds of Exile and Illusion, page 297

The disturbances grow as Ramarren comes to realize that the Shing lie as easily as they speak. They even lie telepathically, as if their very being is a lie. They claim that they are humans whose forebears dressed themselves in the myth of an alien superpower in order to unite humanity and end, or prevent, a great war. But they have destroyed so much—every city, school, library, museum, network—that there is nothing to compare their lies to: every lie they tell exists on its own, unquestioned and unquestionable, not true but not false either. Isolated in time and space, with no referents to knowledge or history, the lies become a kind of anti-truth that, on contact with the truth, annihilates them both.

Among the Shing are humans known as “toolmen.” As children, the toolmen were identified as having “subnormal minds” and brought to the city to be plugged in to the psychocomputers, raised to become servants to the Lords, as the Shing are known. Mute and compliant, they seem a mere extension of the machines they operate: flesh-and-blood computers completing tasks to fulfill their program, with no thought or belief or desire to get in the way. A toolman is the perfect slave: he has no self to rail against his enslavement.

Ramarren seems at first to be treated as an honored guest by the Shing. He is given the comforts a body needs, and spoken to with a formalized, if stilted, respect. But he soon learns that the Shing have twice assaulted his mind, have hunted through it in search of the secret location of his home planet, as if his mind were nothing more than an encrypted storage device, a resource to be hacked and exploited. The Shing’s appetite for the acquisition and destruction of knowledge extends not only to his people but to his own thoughts. And their lies (their anti-truths) are their greatest weapon.

Ramarren—alone and unarmed—has but one small hope against their methods:

Against them he could never prevail except, perhaps, through the one quality no liar can cope with, integrity. Perhaps it would not occur to them that a man could so will to be himself, to live his life, that he might resist them even when helpless in their hands.

Le Guin, Worlds of Exile and Illusion, page 335

Integrity seems a meek defense against the Shing’s aircars and psychocomputers, the parahypnotic drugs, the gravity-defying city and deadly laser guns. And yet that integrity is what allows Ramarren to come back to himself, to finally see through the Shing’s lies, to briefly and decisively disarm them and make his escape. It is in his unremitting longing for life that Ramarren is finally free.