The robots may or may not steal our jobs. Headlines frequently announce that AI will leave half the population unemployed, but a closer look at the data indicates a smaller impact. And history, including very recent history, suggests that automation ends up complementing non-automated work, rather than replacing it.
But even if the robots don’t steal our jobs, they may be stealing our humanity.
By that I mean that the common habit of treating robots as people—however much it seems like a joke or a shorthand or a metaphor—could erode our already diminished sense of the human person as uniquely special. Effusive media reports give the impression that robots will befriend us, play with our children, and care for us when we are old—and that they may be (at least) as good as the human alternative.
“Could Your Child’s Best Friend Be A Robot?” asks Forbes, introducing us to Cozmo, who “loves stacking cubes and playing games with you. If you beat him too many times in a row, he’ll make disgruntled noises, scowl at you, and might even flip a cube in anger.” According to an excited story in the Independent, two Facebook-created robots spontaneously began “chatting to each other in a strange language only they understood.” (A representative extract from this astonishing moment of invention: “ALICE: balls have zero to me to me to me to me to me to me to me to me to.”)
Robots are said to “understand” ideas and “respond” to people, with whom they “form relationships.” They have “ambitions” or “want to help you” (though they may, more worryingly, turn out to be racist). Cosmopolitan recently announced on its cover: “He’s smart, funny and wifi-enabled: Why you will date a robot in 2018.”
These claims help to dehumanize genuine friendship and creativity—and they are often untrue. The author of the Cosmopolitan story, Clare Thorp, in fact demonstrates that you will not date a robot in 2018, nor at any point in the foreseeable future. Google Home and Amazon’s Alexa, like the other robots Thorp meets, are unable to “understand, or help” with emotional issues. The £20,000 Pepper, whose manufacturers allege that it can “read human emotions,” responds to the admission “I’m sad” with “That’s good information.”
When not giving scripted answers on TV programs, the robots are out of their depth. Hiroshi Ishiguro thinks the robot he has created, Erica, has “the most beautiful face in the world.” Ishiguro has begun to “install emotion and desire” in her. He observes, for instance, that she “wants to be well-recognised” in her role as “a kind of receptionist for this laboratory.” But Erica is incapable of even mimicking understanding: A simple inquiry provokes first silence, then a flurry of programmed responses, such as “Are you making fun of me because I am a robot?”
Even if the robots could be designed to respond more convincingly, their essential inadequacy would remain. Human beings are able to form universal concepts based on our experience, and then put those concepts together. Robots cannot begin to emulate this. As the philosopher Raymond Tallis points out, a clock can tell us the time, but that doesn’t mean the clock “knows” what time it is. Robot-chatter is merely a far more complex and multi-layered instance of the same illusion.
A century ago, a horse named Clever Hans went on tour and showed off his mathematical brilliance to admiring crowds. “Hans is an expert in numbers,” the New York Times explained in a 1904 writeup. “When asked how many 3’s there are in 7 he stamps down his foot twice and for the fraction once.” It turned out that Hans was just sensing his owner’s body tension and working out when to stop stamping his foot. When the horse couldn’t see anyone, the depths of his expertise became sadly apparent.
Our electronic Clever Hanses could fool us into half-believing that the human capacities for love, understanding, and virtue—or for wickedness and stupidity—are no different from the kind of thing a computer program comes up with. Robotization is a natural sequel to dehumanization: A pornographic culture is already falling for “sexbots,” and employers who treat factory workers as disposable will think nothing of replacing them with machines.
So the real threat to jobs may be not technology, but the attitude we bring to it. According to Jamie Bartlett, Silicon Valley’s entrepreneurs believe AI will take away tens of millions of jobs. They “don’t say it in public of course, because what’s the point. It’s inevitable, they say; technology can’t be stopped. It’s a force of nature.” How comforting to believe that a process that would enrich you and immiserate others is just part of the great wave of history. This kind of fatalism treats people like pieces of metal that can be thrown on the scrapheap. And it may begin with treating pieces of metal like people.
Dan Hitchens is deputy editor of the Catholic Herald.