By Keith Johnson
Think your job is safe against automation? Think again. Up until recently, the introduction of robotics and computers into the workplace has primarily posed a major threat to low-wage earners who carry out tasks usually on assembly lines. However, as the world marches headlong into the second decade of the 21st century, technology is advancing to such a degree that it is now encroaching upon occupational skills that used to belong to humans alone.
As part of an ongoing series, this newspaper previously reported on this disturbing new trend in the August 1, 2011 issue (#31) on page 7.
According to Massachusetts Institute of Technology scientist Andrew McAfee, “The list of things humans are demonstrably better at than computers is shrinking pretty dramatically.”
In his new book Race Against the Machine, McAfee and co-author Erik Brynjolfsson cast a dire outlook on future employment prospects as human workers become increasingly obsolete.
“The [rapid] pace and [large] scale of this encroachment into human skills [are] relatively recent and [have] profound economic implications,” write the authors. “Perhaps the most important of these is that while digital progress grows . . . it can do so while leaving some people, or even a lot of them, worse off.”
According to these authors, corporate spending on technology has increased 26% since 2009 while payrolls have remained essentially flat in that time period. This has contributed dramatically to the disparity of incomes between rich and poor.
“Corporate profits as a share of GDP are at 50-year highs,” they write. “Meanwhile, compensation to labor in all forms, including wages and benefits, is at a 50-year low. Capital is getting a bigger share of the pie relative to labor.”
Citing economist Emmanuel Saez as a source, the authors go on to say that the top 1% of United States households—that is, the 14,588 families with income above $11.5 million—got 65% of all economic growth since 2002 and saw their share of national income double from 3% to 6% between 1995 and 2007.
McAfee and Brynjolfsson aren’t the only ones sounding the alarm. In the latest edition of the McKinsey Quarterly, renowned economist W. Brian Griffith argues that automation, digitization and robotics is creating a vast and invisible “second economy” that will reach the size of the physical economy within two decades.
“In the early 20th century, farm jobs became mechanized, and there was less need for farm labor,” writes Griffith. “Some decades later, manufacturing jobs became mechanized, and there was less need for factory labor. Now business processes—many in the service sector—are becoming mechanized, and fewer people are needed, and this is exerting systematic downward pressure on jobs.”
McAfee and Brynjolfsson make the case in their book that rapid advancements in artificial intelligence (AI) technology is posing the same threat to median-income, white-collar professionals as more primitive kinds of automation did to blue-collar laborers in the 20th century.
At a recent technology conference in Tucson, Arizona, the Los Angeles Times quoted Brynjolfsson as saying that approximately 60% of the U.S. workforce is engaged in information-processing tasks.
“It’s hard to think of any of those that won’t be profoundly affected and possibly eliminated by these technologies,” the Times reported.
There are still some high-paying professions that are—for the time being—impervious to technology, according to McAfee and Brynjolfsson. Among them are doctors, therapists, managers and sales associates who rely on intuition and complex communication skills to make diagnoses, solve problems and influence behavior on those with whom they interact. Creativity is another domain where humans still maintain the high ground, say the authors. At the opposite end of the spectrum, the authors say that jobs like gardening, busing tables and styling hair are still safe from encroachment by machines only because technology has not yet been able to mimic the dexterity of humans.
However, when taking into consideration some of the most recent advancements in AI and robotic technology, one wonders if there will come a day when machines outperform humans in virtually every way.
What Will New Generation of Robots Be Used For?
By Keith Johnson
At Harvard University, scientists working from funds provided by the Pentagon’s Defense Advanced Research Projects Agency (DARPA) have created a soft rubber-like robot that is more flexible and maneuverable than any existing metallic model. Made of an elastic polymer that is similar to the caulk used to seal bathtubs, the human-shaped device mimics the flexibility of an octopus.
Having no hard internal skeleton or sensors, the robot moves according to the volume of air injected into its four legs, called pneu-nets. It can walk, crawl, slither and even be deflated to a width flat enough to navigate under a door.
The scientific teamworking on this project hope to soon incorporate visual capabilities so that it can one day be deployed on military reconnaissance and search-and rescue missions.
An even more astonishing development in the field of robotics is Qbo: a mobile artificial intelligence device that resembles R2-D2 from the Star Wars movies. Not only is it capable of moving about on its two wheels, but it can also interact with human beings through a sophisticated series of audiovisual features that are hardwired into its chassis.
In a video demonstration, Qbo hears the voice of an approaching technician and identifies him by name after processing the man’s sound patterns through its speech recognition software. Qbo is then shown a picture of a penguin, which it also identifies correctly after the image is captured by its stereoscopic vision lens and then matched against a catalog of images stored on the unit’s software. Though not part of the demonstration, Qbo’s makers claim the software is also capable of recognizing individual human faces so long as the data is prewritten into its memory banks.
The video goes on to show Qbo learning to become self-aware. After the technician tells the device to “take a look around,” Qbo soon finds itself staring into a mirror for the first time. Initially, Qbo is unable to recognize itself. But when the technician tells it, “This is you, Qbo,” the robot logs the image into its memory and then replies: “Oh. This is me. Nice.”
It’s not hard to imagine ways this robot could be put to use by law enforcement agencies like the Department of Homeland Security. We may soon find it patrolling the streets of America—in shopping malls or airport terminals—scouting for faces and voices belonging to those whose names end up on dreaded terrorist watch lists.
We may also find a friendlier version of this robot behind sales counters, where they will interact with customers in a more familiar and pleasant way than the human cashiers they will one day replace.
Keith Johnson is a writer based in Tennessee.