by Madeline Patton
Presentation for the NSF ATE, Preparing Technicians for the Future of Work
December 12, 2018
Mehran Gul, Lead for Digital Transformation Initiative
World Economic Forum, Geneva, Switzerland
At the inaugural meeting of the Preparing Technicians for the Future of Work project, Ann-Claire Anderson, principal investigator of the Advanced Technological Education project funded by the National Science Foundation, noted that “technicians are at the center of technological disruption,” but have so far not been the focus of the many scientists and consultancies looking at workplace changes.
To provide a global context for the ATE Leadership Caucus and the project’s 10-member industry advisory board, Mehran Gul, lead of the Digital Transformation Initiative at the World Economic Forum in Switzerland, gave a 30-minute presentation about the digital transformation of industries in Alexandria, VA. He also answered questions from the audience for 30 minutes afterward.
Historic Predictions vs. Reality
Gul began by sharing headlines from the past 100 years that predicted dire job impacts from machines. Given that those predictions did not match the realities that unfolded, he is skeptical about current predictions of automation eliminating jobs. Instead he expects “augmentation of human capabilities rather than substitution.” Referring to data on total employment growth since 1939 as evidence of “unequivocally upward progress,” he said the number of people employed and productivity have increased with the introduction of more sophisticated machinery in workplaces.
Sharing a new twist on the necessity being the mother of innovation adage, Gul said, “Invention is the mother of necessity. The more that we develop technologically, the more that creates needs that create more jobs that create more needs that create more jobs.” He observed that it’s easy to think about jobs that will be eliminated but difficult to think about what new jobs will be created by new technologies. It’s harder still to predict where the jobs will be. Historic data, however, indicate there will be more jobs from technological innovations.
“If you look at the past half century or so jobs have not become obsolete as a result of technology. And if you really think about it, the entire point of technological progress is to eliminate jobs that we find uncomfortable or undesirable for most people, and to really elevate humanity to do things that we actually like to do,” he said.
The “reallocation” of people to different parts of the economy in response to technological changes can be quite positive. For instance, he noted, in 1900 48% of the U.S. population worked in agriculture. Now just 2% of the U.S. population works in agriculture but produces more food than in 1900.
As an example of the unpredictable nature of automation’s impact he reminded the audience that ATMs were expected to replace bank tellers when they were introduced in the 1980s. The cash dispensing machines did drive down operating costs, and also contributed to banks’ expansion in more places. Fewer tellers staff each location, but there are thousands more tellers than before; their jobs involve forging relationships with customers and doing less routine tasks. Rather than the workforce shrinking, “people got redeployed in jobs they actually wanted to do,” he said.
Computer advances mean that routine jobs can be automated and fit into algorithms. But jobs that require manual dexterity, creativity, strategizing, and abstract thinking have not. So far the jobs of plumbers, hairdressers, and cooks have continued while the jobs of payroll clerks and law clerks have not. And while machines carry out routine tasks very quickly, they are not yet capable of non-routine tasks that require interpersonal interactions, adaptability, and common sense. For example robots currently install windshields on new cars in controlled factory environments, but it still takes a human to replace a broken windshield out in the world. So far computers are not as good as humans at operating in changing environments and doing unpredictable tasks like cleaning up shards of glass around the frame for it seal properly.
The reality that humans know more than they can describe and program into a computer, which is known as Polanyi’s Paradox, is a big issue in artificial intelligence. If scientists can overcome this paradox, and figure out how to teach computers tacit knowledge, Gul predicts a wider swath of jobs will be automated.
O-Rings and Educators
“Tasks cannot be substituted by automation; they are generally complemented by it,” Gul said, pointing out that few current jobs do not involve computers in some way, but rarely do computers operate without human interaction.
He used the Challenger explosion, which was caused by the failure of a booster rocket O-ring that ossified in the cold temperatures the night before the launch in 1986, to illustrate that “any big system is only as strong as its weakest link.” Conversely, he said, “improvements in the reliability of any given link, improves the value of improvements in every other.
“In the future, humans are going to be the O-ring of any job or of any workplace,” he said, noting that society can’t afford for “the vital human link to be the weakest link in the system.” Therefore, the introduction of technology in any workplace means that more attention must be paid to training of humans who use it, and will likely be more important in the future.