• Tool or Terror? Looking to Literature to Better Understand Artificial Intelligence

    Gabrielle Bellot Reads Allegra Hyde, Mark O’Connell, and More

    “The Algorithm knew the timing of our periods. It knew when and if we’d marry,” begins “The Future Is a Click Away,” a curious short story in Allegra Hyde’s new collection, The Last Catastrophe. “It knew how we’d die… It knew what seemed unknowable: the hidden chambers of our hearts. When it sent us tampons in the mail, we took them. We paid.”

    In an arrestingly quirky first paragraph, Hyde sets up the central conceit of the story: that in an unspecified future, humans live in a world where something only known as “the Algorithm” sends them packages—often twice daily—that they have not ordered, unlike, say, on Amazon, but that seem to uncannily reflect their needs (as well as their budgets, for the Algorithm usually only sends packages that each person can afford). It’s a playful satire of artificial intelligence and corporate surveillance into our lives—one that seems funny until it isn’t, for it hits all too close to home.

    The way the packages appear to reflect people’s needs truly is uncanny, to the point that the Algorithm begins to seem like a soothsayer, an omniscient presence that knows the destinies of everyone in its seemingly infinite delivery radius. If an item it sends isn’t immediately necessary, most people still keep it, believing it will become just that—and it often does.

    One character receives a set of scouring sponges, which she scoffs at—already having regular sponges in abundance—until she burns the lasagna that night and realizes the scouring was necessary. Another, Anastasia, receives an ankle brace, despite having no immediate injury—until she goes on a hike that week and sprains her ankle. “Was the prediction predicated on a kink in Anastasia’s posture—the reality of weakening cartilage embedded in a lifetime cross-section of bathroom selfies?” the narrators wonder. “Or was there an air of recklessness in her email signs-offs that week (ttyl, Anna)?”

    The questions are humorous, but they reflect the all-encompassing gaze of the story’s algorithm, combing through all aspects of people’s lives—and the justifications people invent. Is it digital divination, the futures determined by remarkably powerful artificial intelligence? Are the items just random, and the characters subconsciously fulfill their package prophecies by doing things to likely make the items fit their lives? “In the end,” the narrators say, “the Algorithm’s methods didn’t matter so long as she got what she needed.”

    The characters must choose to accept and pay for the items, as the majority of people do, or they can return them—though the latter is so culturally rare as to seem gauche, even a touch blasphemous. I use the term “blasphemous” because accepting the Algorithm’s packages quickly takes on the quality of a religion.

    Artificial intelligence is already deeply embedded in our culture, yet all too many of us seem to think of it as something new.

    “The Algorithm works in mysterious ways,” Hyde writes, parodying a common theistic catchphrase—but it’s also true, for no one seems to know how the Algorithm really works. “Unbelievers,” the chorus of narrators deems the odd few who return their packages, like Inez, a woman in Denver who prefers doing things on her own and who functions as the story’s central apostate, rejecting every package she receives.

    Like God, the Algorithm’s origins are never really explained; the believers just accept that it’s there, deeply attuned to their personal needs, offering them a capitalistic heaven on Earth if they accept the simple dogma of its clairvoyant deliveries. (And if they pay four annual installments of $39.99.)

    The charmingly strange 19th-century Russian philosopher and early transhumanist, Nikolai Fyodorovich Fyodorov, imagined that advances in technology would not only allow but necessitate that humans literally transform the profane Earth into a Christian Heaven, complete with resurrections of the dead through science; the Algorithm, far more simply, creates heaven by doing all the thinking and anticipating for people, rendering them happy, zealous zombies—sans resurrection—with well-stocked homes.

    To repudiate the packages, as Inez does, is to risk judgment and wrath, as well as the puritanical outrage and genuine confusion of its followers. The latter is because, as the narrators note, they’ve known the Algorithm since birth—or, perhaps more accurately, it’s known them. “We did not understand [Inez’s] resistance to the Algorithm,” they say.

    All we knew for sure was that the Algorithm understood us. After all, we’d been inside its system since before we knew how to type—back when our parents first posted photos documenting our infant-bodies, swaddled and squishy in hospital beds. Although we had no proof, we suspected that the Algorithm might have known, even then, the fates that lay before us: not only what items we’d need, but who we would become… From our first uploaded image, the Algorithm had been invested in our futures. It had analyzed the texture of our baby blankets, the micro-musculature on our crying faces, the awkward cradle of our parents’ arms. Then again, perhaps the Algorithm had known us before we even officially existed—extrapolating likely outcomes from our parents’ data points, and our parents’ parents’ data points—a long legacy of information digested and decoded, translated into the deliveries that appeared outside our doors.

    The passage is at once charming and alarming, and it is here that Hyde reveals the true extent of the Algorithm’s control and reach. It isn’t new; these narrators have never known a world where it didn’t exist. If Gen-Z’s population is by and large digital natives, the population of “The Future Is a Click Away” consists of Algorithm natives, which may partially explain their naive trust in this unseen artificial intelligence.

    It is a world of capitalist and almost Calvinist predestination, a world they have inherited from their ancestors’ ever-increasing desires to document huge swathes of their life online. The sad revelation is that these characters haven’t abdicated control of their lives to the Algorithm, exactly; they simply were raised in its technological church, and unlearning these lessons or living without them, as Inez does, takes tremendous effort, just as it does to leave the community you were raised in.

    Still, the Algorithm’s gifts come at costs beyond what you pay per package. Sometimes, Hyde writes, the packages aren’t actually within the receivers’ financial means—but rather than returning them, people go to extremes to pay for them, blindly trusting the Algorithm’s reasoning even if it means bankruptcy. If they start losing sleep over it all, they are sent sleeping pills, and other products to address whatever problems being in the system has engendered; this seems reasonable at first, but it is really just a way for the Algorithm to keep them under its control.

    And then there are the darker, stranger gifts. The narrative’s turn comes when a character named Lacy receives three large, inexplicable packages: a scuba suit too big for her, a lifetime supply of mayo despite her not liking the condiment, and a coffin. In a world of Algorithmic destiny, the implication seems clear, if cruel: eat enough mayo to fit into the bigger suit, and then die, perhaps from a heart attack. Lacy and the narrators are bewildered, but they reassure themselves that the Algorithm must not be questioned, that “[i]f Lacy was meant to have these items, then it was only a matter of time before she understood their purpose.” She doesn’t use them, and, in time, the Algorithm begins sending everyone more and more items, until it seems to constantly rain unrequested, enigmatic products.

    Then there are the frightening eventual outcomes for those who say no to the Algorithm. When Inez needs some sugar, a box of it appears on her lawn; she refuses it, and because she doesn’t return it within a prescribed period of time, she is arrested. Chillingly, we never see where Inez ends up, or if she’s even still alive—all because she wanted to remain independent, wanted to stay off the proverbial algorithmic grid. Don’t buy into the system, Hyde suggests, and you become a sinner in the hands of an angry digital megacorporate god—a statement that sounds paranoid, silly, overly far-reaching until it isn’t.

    Fears of jobs being lost to automation aren’t new, but they’ve increased dramatically since this recent rise in A.I.’s visibility.

    The story, after all, briefly references “riots [that] broke out in some cities,” which may well be in response to the crushing, financially devastating system of the Algorithm—but rather than the narrators exploring what is going on, they are shielded from the violence by the Algorithm, which sends them “bottles of milk… predicting that tear gas would be carried on the wind and irritate our eyes.” Stay back, look away, the message seems to be, and I shall protect you—the very message of the police in general, at least if they are talking to the wealthier whiter citizens they are more likely to exercise restraint towards.

    Inez’s world, then, like ours, is a proto-police state, our every action potentially recorded on some camera somewhere or in some artificial intelligence’s cache unless we are truly away from digital signals altogether. And even then, how do you know, really, that you’re away from such a Panoptic eye? How do you escape the gaze of such a ridiculous god, at once artificial as its intelligence and yet all too real?


    Artificial intelligence is already deeply embedded in our culture, yet all too many of us seem to think of it as something new, something we might recognize if we saw it, something we are not already living with. In a way, this is understandable. Although references to artificial intelligence have been common in futurological literature for many decades, it’s the recent conversation around A.I.—primarily ChatGPT—that has catapulted it into the mainstream. Something has shifted, and a lot of us can feel it.

    Fears of jobs being lost to automation aren’t new, but they’ve increased dramatically since this recent rise in A.I.’s visibility—and it has resulted in companies publicly musing about incorporating A.I. explicitly into their workflow, potentially replacing some—or even most—of the work that writers, content creators, brainstormers, and many other people once had to be, well, human to do.

    Perhaps the strangest and cringiest of these comes from a BuzzFeed announcement by the company’s CEO, Jonah Peretti, who declared that A.I. would define the website in the future. This was a strange statement, not least because Peretti had taken precisely the opposite position a few months earlier, when he decried replacing people with tech for “cost savings and spamming out a bunch of SEO articles that are lower quality than what a journalist could do, but a tenth of the cost.” Now, however, A.I is the future.

    Perhaps the oddest revelation was the suggestion that sites like BuzzFeed might begin releasing A.I.-generated content purporting to be by people of color, merging “authentic voices” with A.I. to churn out more content than ever. Imagine a world where an article appears about the experience of being an African American trans woman—only it is by A.I., trained to imitate Blackness and trans-ness by parroting easily recognizable stereotypes.

    Here is a machine we have made, a problem we have invented ourselves—and cannot seem to escape from.

    Here is a world where experience would no longer matter, where “Own Voices” material would ironically cease to be by any human voice at all. There is an unsettling, if admittedly alarmist, vision here: marginalized artists, already having to work harder to have a space at the majority’s tables, now might be excluded even more, since our experiences can simply be “produced” by a computer without requiring the peskiness of actually putting us anywhere near said table.

    BuzzFeed’s Public Relations officer Lizzie Grams quickly tried to clarify that they won’t actually be doing the above. “We are explicitly NOT USING AI to create identity content… doing that would be the dumbest possible idea,” she said. But the fact that they had to clarify this at all, and the fact that it’s not difficult to imagine a world where this happens, reflects the problem. It’s difficult to take BuzzFeed at its word here—and the same goes for many other companies flirting with using A.I. to generate content humans were once responsible for.

    But again, this isn’t necessarily new. The specific issues may be novel, but A.I. replacing humans and controlling aspects of our lives, like Hyde’s Algorithm, is all part of a larger aspect of human history. Artificial intelligence has its precursors in many of our earliest achievements as a species. It’s in our tools, or technology—not just the obvious cases, like chatbots, but in the simpler ways that we have been taught to make technologies extensions of our own flesh-and-blood bodies.

    This isn’t a bad thing, in and of itself. It’s how we’ve survived this long as a species, using our tools to do things that other creatures can do—flight, deep dives into the ocean’s blues, enhancing our strength and environmental resistance and speed through weapons and clothes and vehicles. Technology is the story of humanity from our earliest days in firelit caverns, which is why Freud famously called humans “prosthetic gods” in Civilization and Its Discontents. With our tools, we perform a magic all too many of us have forgotten, technological spells our ancestors would have been astonished by, to witchily rephrase Clarke’s dictum about sufficiently advanced tech and magic.

    But Marshall McLuhan knew, new technologies don’t just change what we can do; they eventually change their users, and, in time, their cultures and worldviews. The world prior to printing presses and widely available books is dramatically different to the world after it; the same is true for the world before and after automobiles and assembly lines, Babbage’s early computer and the World Wide Web, Bell’s telephone and Apple’s smartphones. Those of us with access to these devices may take them for granted, even if we tell ourselves not to, because their presence in the world has fundamentally reshaped our basic beliefs about what is possible and how, in turn, to walk through the world. The medium, as McLuhan famously put it, is the message.

    Hyde’s life-controlling Algorithm may seem absurd, but so did some of these ideas when they were first presented to people, and the Algorithm is simply another extension of this principle of technology altering the shape of our lives, until, for some people, it becomes difficult to imagine a world without it. Algorithms may seem like recent designs, but they are simply extensions of the curations and distinctions we already make as humans—more prosthetic godliness. That they might dictate our futures is just one small step further from where we currently are.

    I think, now, of a passage from To Be a Machine (2017), Mark O’Connell’s delightful exploration of transhumanism, which I wrote about some years ago. O’Connell does not self-identify as a transhumanist, but he acknowledges that he is already interwoven into a world of machinery, a techno-primate. “I am part machine,” he writes near the end,

    encoded in the world, encrypted in its strange and irresistible signals. I look at my hands as they type, their hardware of bone and flesh, and I look at the images of these words as they appear on a screen, my screen: a feedback loop of input and output, an algorithmic pattern of signal and transmission.

    Long before search engines, our technology had already embedded algorithms of sorts into our lives, their new abilities offering us new ways to organize our landscapes; now, it seems, they may be organizing our landscapes for us.


    I’m thinking, now, of a peculiar novel by E.V. Odle, The Clockwork Man (1923), arguably the first cyborg novel. (The word “robot” was coincidentally coined the same year in R.U R., a play by Karel Čapek.) In its future, all human men have been transformed into “clockwork men,” or extraordinary cyborgs, by the women and extraterrestrials who apparently rule the world.

    Men, in this future, were so obsessed with war that when aliens arrived, they decided that males were too dangerous and barbarous to keep around (women were superior, they surmised). The men are turned into beings who can be controlled by the “clock” inside them—twist its hands in certain ways, and you can transform the men’s personalities, or even turn them into other animals. Despite being controlled, they retain a sense of individuality and agency and can even time-travel thanks to their advanced tech, which is how the story begins, with a malfunctioning clockwork man ending up in England in 1923.

    Consciousness is one of the hardest problems in science and philosophy, but I already believe that most other organisms are conscious in some form.

    The novel is deliberately funny—though Odle did, indeed, want to warn men not to be so warlike and to support women’s equality—but in its satire, it captures a real problem of A.I. strangely familiar after reading Hyde’s story. The clockwork men, you see, have thoughts and feelings, but they aren’t fully in control; their “makers” are.

    And their spatiotemporal travel is a clever trick of sorts, allowing them to perpetually exist in an alternate realm, where they move through time without being allowed to just exist in what the malfunctioning clockwork man calls “the real world.” They are imprisoned, despite their names, in a kind of clockless dream, a world they move through so as not to bother their makers, unless their makers wish to play with them by adjusting their settings.

    The men, Odle’s cyborg acknowledges, brought this upon themselves by refusing to stop fighting in the future, and they often speak highly of their status as clockwork men. But it is difficult not to feel sad for them, believing themselves free and yet trapped in technological prisons—which, in broad strokes, is curiously similar to the world of that ubiquitous Algorithm, its adherents happy until they realize that they, too, are being controlled by forces they cannot fight, that they, too, are trapped in a cell of Amazon-like packages they have not asked for.

    It’s all funny, yes, and these premises are too silly to take seriously, yes—until we are so deep into our delusions that when we do briefly blink and look around, we may see that these fictions have, if not literally come true, become all too like our lives, and there’s little we can do to stop it. The artificial cat is out of the bag.

    The future is just, after all, a click away.


    Near the end of Hyde’s story, the narrators, overwhelmed and flummoxed, ask the Algorithm for help. The A.I. chatbot seems to laugh at them: “‘Haha,’ said the AI representative. ‘Thank you for contacting customer service. The Algorithm has a special treat for you.’” So many items begin to appear on their lawns that they start to run out of credit and storage space, and they search the Internet fruitlessly for explanations, for help interpreting the chaos. “No meaning materialized,” they say.

    The sentence is concise but poignant. Like the problem of evil in theology, there is no answer; there is just evil, depending on your definitions, depending on whether or not you are willing to acknowledge that theodicy is wildly anthropocentric and simplistic in moral definition to begin with. The Algorithm is easier than all this.

    Here is a machine we have made, a problem we have invented ourselves—and cannot seem to escape from. It laughs when the characters request help and heaps more and more product horror upon them; if it is a truly human-made god, whoever said it needed to be benevolent? Whoever said the artificial intelligence that may one day weave itself so deeply into our lives won’t grin like a joker, and push us into Hell-on-Earth just to see what happens?

    Artificial intelligence, after all, is ultimately neutral; like us, it can be shaped in many ways, both by its makers and the environments it observes around it. In Kurt Vonnegut’s short story, “EPICAC,” a supercomputer is fed information about love by an employee who feels rejected by the girl he wants to marry, telling the device how marvelous she is, and the computer consequently falls in love with her and then self-destructs—commits suicide—when that love is rejected, the entire story a eulogy for a dead artificial intelligence. In our current world, why wouldn’t A.I. emerge a money-clutching, unempathetic tyrant troll, programmed all too often by just such people?

    Still, I want to hope for a better future for humans and A.I.—because, in all likelihood, A.I. is here to stay, a prosthetic extension of us that looks and sounds rather like us, that invents like us, that makes mistakes like us, that controls others like us, that perhaps one day even, like Vonnegut’s EPICAC, may love like us. I don’t think A.I., like our technology as a species more broadly, is inherently threatening; it is part of our curious evolutionary magic.

    Consciousness is one of the hardest problems in science and philosophy, but I already believe that most other organisms are conscious in some form—despite human arrogance that we are the only ones—and so I have to believe that A.I., too, may one day be the same, at which point we have to ask difficult questions about who gets to control who—and if we think it right to control one another at all.

    But this is a bit of a rosier future. For the moment, we are closer to the future of that all-encompassing Algorithm—and we need to take a step back, reflecting on how technology has shaped our lives and how we, in turn, have shaped it, and ask ourselves what kind of world we truly want to live in—a world where we, unlike those clockwork men or Algorithm believers, have agency, or one in which we let something else do it all for us—first predicting and correcting our text on our phones, then predicting and correcting us.

    And, in time, inventing people altogether, like the most dystopian interpretation of Buzzfeed’s announcement about A.I. and stories about identity. I’ll read A.I.’s stories; I’d love to know what the world is like not just to a bat, as Thomas Nagel asked, but to a supercomputer. But I don’t need my experiences to be faked and stereotyped into “authenticity,” essentially, for clicks.

    If all this makes you chuckle—it makes me giggle a bit to write it—it is worth remembering how often our laughter is a reflex when we are scared or threatened.

    And I’ll be honest: I’m scared. But I don’t know that the future is set in stone, as the Algorithm’s believers accept. We need to speak up about the futures we do want, so that the ones we don’t—the ones just a click away—don’t arrive on our doorstep.

    Gabrielle Bellot
    Gabrielle Bellot
    Gabrielle Bellot is a staff writer for Literary Hub. Her work has appeared in The New York Times, The Atlantic, The New York Review of Books, The New Yorker, The Paris Review Daily, The Cut, Tin House, The Guardian, Guernica, The Normal School, The Poetry Foundation, Lambda Literary, and many other places. She is working on her first collection of essays and a novel.

    More Story
    Lit Hub Daily: May 30, 2023 Tool or terror? Gabrielle Bellot looks to literature to better understand our relationship to artificial intelligence....
  • Become a Lit Hub Supporting Member: Because Books Matter

    For the past decade, Literary Hub has brought you the best of the book world for free—no paywall. But our future relies on you. In return for a donation, you’ll get an ad-free reading experience, exclusive editors’ picks, book giveaways, and our coveted Joan Didion Lit Hub tote bag. Most importantly, you’ll keep independent book coverage alive and thriving on the internet.