agar3_gettyimages_humanrobotheadsline Getty Images

Automating Segregation

As more companies seek higher profits through automation, artificial intelligence, and other digital innovations, we should consider what is lost when we replace humans with machines. The fact is that we need other humans in our lives to avoid despair, and we need encounters with those unlike ourselves to sustain our diverse democracies.

WELLINGTON – There are many legitimate complaints about work in the “gig economy.” But, in time, many of the gigs will be automated, eliminating a significant source of work for immigrants and the less educated. And as these jobs disappear, so, too, will social contacts between socioeconomic classes, and between immigrants and natives. The broader question, then, is the impact of the impending loss of involuntary social mixing on diverse, multiethnic societies.

Consider Uber, which provides work to many immigrants. According to the company’s website, Uber is keen to foster “happiness and inclusiveness” by means of “global cultural and community events to increase and enhance cross-cultural learning and understanding.” But while that people-centered message certainly sounds nice, Uber has placed its hopes – and its future profitability – in automated vehicles (AVs).

In anticipation of its disappointing 2019 initial public offering, Uber disclosed losses of $1.8 billion for the previous year, owing partly to its payout of $1 billion in driver referrals and other incentives to gig workers. Since the IPO, the company’s stock has struggled, because it has yet to demonstrate that it can turn a profit with its human-driver model. AVs are an obvious potential solution to this problem, as Uber’s founder, Travis Kalanick, foresaw back in 2013. Viewing a Google driverless-car prototype, he concluded that, “The minute your car becomes real, I can take the dude out of the front seat … I call that margin expansion.”

Needless to say, gig work in the “sharing economy” has disappointed earlier expectations that it would produce an explosion of micro-entrepreneurship. Driving an Uber is no one’s dream job, and the company’s drivers have found themselves locked in a protracted battle for better pay and work conditions. Nonetheless, we should bear in mind the social interactions that will be lost when all Ubers – indeed, all transportation – become driverless.

The same concerns apply to retail work, which Amazon plans to eliminate through its Amazon Go cashier-less convenience stores. Again, retail is nobody’s dream job, not least because it doesn’t pay well. But if we eliminate all these positions, what occasions will we have to interact with others unlike ourselves?

Humans are naturally gregarious animals. Our happiness depends on being with other humans. But this fundamental trait combines awkwardly with another feature of our evolved psychologies: suspicion of strangers. This makes perfect sense when one considers that the forager bands of our Pleistocene ancestors had around 50-100 members, many of whom were related.

Secure your copy of PS Quarterly: Age of Extremes

Secure your copy of PS Quarterly: Age of Extremes

The newest issue of our magazine, PS Quarterly: Age of Extremes, is here. To gain digital access to all of the magazine’s content, and receive your print copy, subscribe to PS Premium now.

Subscribe Now

In fact, humankind’s crowning achievement is not the Moon landing or the invention of the computer. It is that we, the descendants of xenophobic hunter-gatherers, have built sprawling, diverse societies comprising tens of millions of strangers. Though many students who attend diverse colleges say they enjoy meeting new kinds of people, research shows that even these social butterflies tend to stick with their own. Only when they join the working world do they find that they have no option but to get along with strangers who look, sound, and act unlike themselves.

But even then, our shy, occasionally murderous inner apes will reassert themselves in times of stress or political and economic uncertainty. In recent years, some of us have found it disturbingly easy to imagine that all immigrants are bringing drugs, crime, and disease into our countries. And social media’s biased sampling of our online expression has aggravated this problem, as well as deepened social, partisan, and other divisions.

How will we form social connections and develop empathy in an economy where teachers, baristas, cab drivers, and retail clerks have all become highly efficient machines? We certainly cannot count on “social” platforms like Twitter, where our murderous inner apes reign supreme.

Let us return to the example of Uber. When you hail a ride, you can act insultingly or simply sit sullenly, but this may affect your rating; in the future, nearby drivers may skip over you for higher-rated passengers. But if you engage your drivers in interested and polite conversation, you will most likely maintain a good rating. Better yet, these incentives generally lead to enjoyable exchanges. You might just learn something interesting from someone you wouldn’t otherwise have met.

As for retail clerks, there has been much discussion about the growing social divide between workers in San Francisco’s flourishing tech industry and all those who perform traditional services for them. Store clerks don’t get to rate Google software engineers as customers. But at least when they come face to face, each gets a glimpse of the other. When an initiative proposing a higher minimum wage or more spending on affordable housing appears on the ballot, those software engineers may be more likely to support it because they can picture whom it will benefit. If Amazon Go eliminates such encounters (San Francisco already has four stores), will they still support social policies that don’t help them directly?

To be sure, commentators in 2030 will probably look back at the introduction of Amazon Go as the beginning of the end of the human store clerk. But as citizens of diverse democracies, we should recognize what is at stake when we use technology to disintermediate economic interactions. There may come a time when we need to subsidize expensive, less “efficient” work, much as we do now with renewable energy. The fact is that we need other humans in our lives to avoid despair, and we need encounters with those unlike ourselves to sustain our diverse democracies. Automating service jobs may solve some problems, but it will create plenty of new ones. Will the tradeoff be worth it?