Everyone knows that robots, perhaps souped up one day by artificial intelligence (AI), are entering the global labor force in droves. Where there's less agreement is what impact these technologies ultimately will have on workers, jobs and the broader economy.
Indeed, as a new Pew Research report makes clear, the answers to that question are all over the map. Robert Cannon, an Internet law and policy expert, bluntly predicts that "everything that can be automated will be automated." And that process has begun in earnest, he notes, with robots and software already replacing everyone from hardware store employees who make keys to law clerks.
So, what unique skills or attributes do people possess that can't be computerized out of existence? "The short answer is that if the job is one where that question cannot be answered positively, that job is not likely to exist," Cannon said.
Tom Standage, digital editor for The Economist, also argued that robots and AI are likely to transform certain occupations much more quickly than previous waves of innovation. That's likely to heighten economic inequality by widening the gap between more skilled employees whose jobs are harder to automate and people in more routine kinds of work, pushing many into lower-paid service jobs. "This is a recipe for instability," he told Pew.
Others -- including, perhaps not surprisingly, some luminaries of the tech world -- are more sanguine about how much robots and AI will affect work. "Historically, technology has created more jobs than it destroys, and there is no reason to think otherwise in this case," said Vint Cerf, chief Internet evangelist for Google (GOOG). "Someone has to make and service all these advanced devices."
One economist also notes that automation and computerization has historically made workers more productive without killing off jobs. In short, he says, those fleets of self-driving cars that seem all but certain to flood our roadways may not necessarily put a taxi driver out of a job.
One reason for the differences of opinion about technology's impact is that both camps are, to a varying degree, correct. The Industrial Revolution kicked off an unprecedented period of economic growth, while subsequent waves of innovation, such as the proliferation of office machines in the late 19th and early 20th century, created new jobs.
In this way, technology hasn't put workers out to pasture so much as shifted the kinds of work available to them. The farm gave way to the factory.
Less clear is what will happen when robots, which are already taking over the factory, enter other professions. Economist and technology theorist W. Brian Arthur thinks technology endangers higher-skilled professions -- including in the high-tech industry itself. For instance, the emergence of "cloud computing" threatens to replace data center and network managers who run corporate information systems. And who needs a statistician or stock trader when an algorithm will do?
A 2013 study by Oxford University researchers concluded that roughly 47 percent of all U.S. workers are at risk of seeing their jobs automated over the next decade or two.
An issue the Pew report touches on only in passing may, in fact, be the most important determinant in how robotics and AI affect the labor market in the years to come: us. After all, innovation doesn't happen in a social and political vacuum -- it's shaped by everything from tax rates and patent laws to labor rules and trade policy.
As the liberal-leaning Economic Policy Institute notes, technological change has been eroding the number of middle-wage jobs since at least 1950. Yet until the late-1970s, all workers -- blue- and white-collar -- enjoyed strong wage growth. That seems to undermine the argument that robots and AI are the chief threat to workers.