The following text is an excerpt from the book “Artificial intelligence. What everybody needs to know” (Oxford University Press, 2016) by Jerry Kaplan.
Reproduced with the kind permission of © 2016 Oxford University Press
While it’s tempting to think of AI systems in general, and robots in particular, as mechanical laborers competing for employment, this isn’t a helpful perspective from which to explore their impact on labor markets. The image of rolling in a robot and walking a worker to the door may be compelling, but it tends to obscure the more important economic effect: automation changes the nature of work. Obviously, technological improvements have raised productivity and increased economic output throughout human history, most notably during the industrial revolution. In plain language, this means that fewer people are needed to perform the same amount of work.
But it’s equally true that historically, the increased wealth resulting from these improvements has created new jobs, though this effect is rarely immediate. More important, the new jobs are seldom comparable to the ones lost, so the displaced workers often lack the skills needed to fill the newly created positions. As long as these effects are gradual, the labor markets can adapt gracefully, but if they are rapid or abrupt, significant dislocations can occur. The history of agricultural employment in the United States is an example of successful labor force conversion. In aggregate, the loss of farm jobs has been nothing short of apocalyptic. In 1870, 70 to 80 percent of the U.S. labor force was employed in agriculture; by 2008 this number had dropped to under 2 percent. In other words, 150 years ago virtually every able- bodied person worked the land, while today almost no one does. Had this happened overnight, unemployment would have been cataclysmic.
But of course, no such thing occurred, because over the intervening century and a half there was plenty of time for labor markets to adapt. People whose primary skills were planting and harvesting died of old age without the need to learn how to type or drive (for instance), while the resulting wealth created substantial new demand for novel goods and services of all kinds, from smartphones to personal trainers. But the actual process by which machines displace human workers is much more subtle. In practice, automation replaces skills, not jobs, and correspondingly, what employers need is not workers but the results obtained by applying those skills. To be successful, makers of robots don’t have to replace people; they have to provide machines with the requisite skills to perform useful tasks. And while their products may not replace workers one to one, they nonetheless put people out of work because fewer workers are required— what makes some workers more productive also puts other workers out of their jobs.
But the process also changes the jobs of those still employed by eliminating the need for a particular skill, and possibly adding requirements for new skills. A good example of this process is as close as the checkout stand at your local supermarket. The clerks who total your bill (cashiers) and package your groceries (baggers) are engaged in a number of skill- based tasks that have changed significantly over the past few decades. The cashiers used to examine each item in your shopping cart and key the price into their cash register, whereas now they simply swipe the items across a bar- code reader. The advantages of the new system in terms of accuracy, time, and labor are obvious. But the main reason the cashiers are still there to greet you is that certain items require special handling. In particular, bags of loose produce need to be identified and weighed to determine a price. And these are skills that so far, at least, have resisted automation. So did this save the jobs of the cashiers? Sort of. They are still employed, but fewer are needed.
The U.S. Bureau of Labor Statistics project that the need for cashiers (in general, not just for supermarkets) will grow only 3 percent over the next decade, compared to overall employment growth of 11 percent, mainly due to increased automation.2 At this time, the baggers’ jobs are more secure, because properly loading a random collection of groceries into a bag so that it isn’t too heavy, is evenly distributed, and does not damage the contents currently requires human judgment. However, their jobs are being threatened by a nearby competitor— the cashiers themselves, who increasingly are usurping this function.3 Nothing about AI changes the fundamentals of how labor markets evolve with technology.
From an economic standpoint, AI technology is just another advance in automation. But its potential to rapidly encroach on current workers’ skills is unparalleled in the recent history of technological innovation, with the possible exception of the invention of the computer itself. Consider what might have been different at the supermarket had the current state of the art in computer vision been available decades ago. Instead of reengineering the whole process of identifying and labeling items with bar codes, newly installed readers might have been fully capable of identifying items by their visual appearance, if necessary reading the prices written or printed on them. Since this approach would have caused far less disruption to the supply food chain (so to speak), it likely could have been adopted much faster, and certainly at lower cost, causing a more rapid contraction in the workforce.
In summary, to understand whether AI is going to put someone “out of a job” it’s necessary to understand what skills, in aggregate, that worker utilizes, whether those skills are separable from the rest of the work he or she performs, and how susceptible those skills are to automation, with or without the application of AI. As a general principle, the fewer uniqueskills a given worker utilizes, the more vulnerable he or she may be to replacement by a machine, depending on the skills, of course. But even if only a portion of a worker’s expertise or experience is amenable to replacement, improving productivity has the effect of reducing overall employment. So yes, robots are going to take our jobs, but a more useful way to think about it is that they are obsoleting our skills, a process economists call “de- skilling,” appropriately enough. And there’s nothing new about this process— the magnitude of the impact of AI in particular will depend on how quickly and widely the new technologies will facilitate automation of workers’ skills. And on that front, the news is not good for humans.