If artificial intelligence replaces every job, humans will face a mass disillusionment. It won’t be the humanist disillusionment we expect — the mythologized anticlimax of humankind, a world where machines outperform us at jobs we thought were uniquely ours. Rather, we will realize that those jobs never needed to exist, and more begrudgingly, that no job needs to exist.
In his now infamous op-ed and subsequent book, “Bullshit Jobs,” anthropologist David Graeber argues that many workers spend their entire working lives devoted to tasks they believe are useless. While Graeber correctly diagnoses what has come to be a phenomenon of the professional class, he overstates capitalism’s role in producing it. If the age of AI reveals anything, it’s that bullshit work is not so much a symptom of capitalism as it is of humanity.
Graeber begins his account with a historical antinomy. He writes that although economist John Maynard Keynes envisioned a feasible 15-hour work week as early as the 1930s, humanity has somehow managed to keep itself busy, 40 hours a week, for nearly a century.
Graeber attributes this largely to the rapid growth of the service and administrative sector, the invention of unnecessary industries and the inflation of existing ones. He argues that these developments produce what he calls “bullshit jobs”: middle-management positions rife with tenuous bureaucratic work.
Though Graeber’s claims are now nearly a decade old, there is good reason to believe them. His ideas have found support in emerging case studies, and become increasingly relevant to a rising class of white-collar tech workers.
In the age of artificial intelligence, his claims materialize in the predicted AI-overhaul of an already superfluous workforce. We might soon realize that the AI craze may ultimately culminate not in the replacement of essential labor, but rather in the automation of jobs that at best never needed to exist, and at worst never truly existed at all.
“We might soon realize that the AI craze may ultimately culminate not in the replacement of essential labor, but rather in the automation of jobs that at best never needed to exist, and at worst never truly existed at all.”
In some ways, this is already a reality — one that tech CEOs have become aware of and capitalized on. Dubbed “AI washing” by pundits, some recent tech layoffs framed under the guise of “advancements in AI technology” have functioned as mere PR stunts to covertly lay off nonessential employees in an acceptable fashion.
Layoffs and the repackaged bullshit of an automated workforce, however, only tell us half of the story. “Bullshit Jobs” is largely a treatise on workplace uselessness, but it also attempts to explain why the phenomenon of bullshit jobs exists in the first place.
Graeber himself blames it on capitalism: he argues that the “ruling class has figured out that a happy and productive population with free time on their hands is a mortal danger” and that the idea that “anyone not willing to submit themselves to some kind of intense work discipline for most of their waking hours deserves nothing, is extraordinarily convenient for them.”
This claim, however, falls flat in an AI era dominated by a winner-take-all race to automate the very “intense work discipline” Graeber speaks of. If the ruling class truly wanted to keep us busy, its many billionaires and one trillionaire would not be pouring this many resources into AI systems deliberately designed to replace the workforce they built to keep us docile.
Instead, even the occupations that may be deemed “useful” and exempt from automation — like being a teacher, a nurse or an electrician — can be just as useless as the administrative claptrap that undergirds them.
In subsequent scholarship on “Bullshit Jobs,” different case studies reported widely different results, with workers’ perceived workplace uselessness ranging from 4.8% to as high as 37% on the same measure. While this divergence can be attributed partly to differences in the surveyed region, it’s also largely based on what questions workers were asked.
Whereas some studies ask participants whether they “have the feeling of doing useful work”, others ask them whether they think they are making a “positive impact on [their] community and society,” with the latter question generally reporting much higher percentages of perceived uselessness than the former across multiple studies.
“Usefulness” need not correlate with real-world impact. The as-yet-speculative AI utopia will force us to come to terms with that understanding in a way we never have before.
If the replacement of jobs ensues, we will, best case scenario, no longer need to work. Agentic systems will optimize all of society’s critical vocations, agricultural planners, infrastructure operators and doctors alike. The workforce as we know it will be abolished, exposed as the largely self-perpetuating system of inefficiency that it is, and society will become seamless.
Instead, critical thinkers, innovators and artists will begin spawning in such abundance that we may no longer even need more of them.
Humanity will then reach their own inflection point just as the machine has, and return to a more animalistic, albeit familiar, form: bullshit. Perhaps we will invent bureaucracies that need not exist or invent questions that need not be asked. In any case, we will be more useless than ever.
According to historian Johan Huizinga, culture largely originates in the human “play” element — an intrinsic drive towards the make-believe that brings rise to law, art, war and the sciences. Perhaps by automating away what was once real, it will be the machine that allows us to play make-believe once again.
Facebook Comments