Robotics and the Economy

September 25, 2015 – Volume 25, Issue 34
Should workers fear increasing automation? By Patrick Marshall

Introduction

A robot serves customers at a restaurant in China's (Getty Images/ChinaFotoPress)  
A robot serves customers at a restaurant in China's Zhejiang Province. Within 10 years, some experts predict, robots and sophisticated computers will be able to perform nearly half of the work done by humans. (Getty Images/ChinaFotoPress)

Robots and other forms of automation can be found not only on factory assembly lines but also in hospital operating rooms, behind car steering wheels, in offices and on Wall Street trading floors. What's more, some experts say, the capabilities of robots and intelligent machines — computers that can analyze data and act on it autonomously — are improving so rapidly that within 10 years machines could perform nearly half of the work done by humans, and more cheaply. Robotics proponents say automation will improve productivity, lower labor costs and make companies stronger and more competitive with overseas rivals. But automation critics fear robots will put many Americans out of work, cause social upheaval and disrupt entire economic sectors. Meanwhile, the legal system is scrambling to catch up with the fast-growing robotics sector, with experts expressing confusion over who should be responsible when robots go awry and cause physical or economic harm.

Go to top

Overview

The hardest-working chef around wears no apron, takes no breaks and is not one for idle talk. Indeed, the Burger Bot can grill 360 burgers per hour, place them on buns, layer them with lettuce, tomatoes, pickles and onions and wrap them in paper.

The prototype burger flipper, according to San Francisco-based Momentum Machines, can do the work of two or three full-time workers, and do it better. “Our device isn't meant to make employees more efficient,” said co-founder Alexandros Vardakostas. “It's meant to completely obviate them.”1

Some restaurants already are so automated that customers never interact with a human. At Eatsa, also in San Francisco, there are no waiters, counters or cashiers. Customers order on a flat-screen monitor, pay on an iPad and pick up the food in a cubbyhole.2

Robots' impending presence in restaurants is one sign of a trend in the U.S. economy: Driven by rapid advances in robotics, automation is spreading to all facets of American life. Machines are diagnosing diseases and performing surgery, flying aircraft and even writing financial reports.

The progress has robot enthusiasts excited over potential gains for the economy and productivity, while some economists and academics worry about job losses and other disruptions to the economy.

At the annual conference in October 2014 held by Gartner Inc., a technology research and advisory company, Research Director Peter Sondergaard said robots or other smart machinery will perform one in three jobs by 2025.3

The PackBot robot, made by Massachusetts-based iRobot (Getty Images/Bloomberg/Michael Fein)  
The PackBot robot, made by Massachusetts-based iRobot, was sent to Japan in 2011 to measure radioactivity at the Fukushima nuclear plant after it was hit by an earthquake and tidal wave. As experts debate the impact robots and smart machines will have on society, some foresee economic gains, including increased productivity, while others worry about job losses and other disruptions to the economy. (Getty Images/Bloomberg/Michael Fein)

“If that's true,” warns computer scientist Steve Omohundro, founder of Palo Alto, Calif.-based Self-Aware Systems, which studies robotic safety, “we're talking massive social dislocation.”

Software entrepreneur Martin Ford, author of Rise of the Robots: Technology and the Threat of a Jobless Future, agrees. “We're moving toward an age where machines are going to take on more and more of the work,” he says.

Robotics proponents dismiss such warnings.

“The fears of economists, politicians and workers themselves are way overdone,” writes John Tamny, editor of RealClearMarkets.com. “Robots will ultimately be the biggest job creators simply because aggressive automation will free … up” capital for investment. “The profit-enhancing efficiencies that robots personify (even to their most ardent critics) foretell a massive surge of investment that will gift us with all sorts of new companies and technological advances that promise the invention of new kinds of work previously unimagined.”4

Other proponents say robotics will allow some companies to bring manufacturing jobs back to the United States because automation lowers labor costs and allows firms to compete with low-wage competitors overseas. “It would be better if U.S. companies were highly automated, were more productive and were producing better quality products for people around the world, so that we could save manufacturing jobs here and create new ones,” says Jeff Burnstein, president of the Association for Advancing Automation, a trade association in Ann Arbor, Mich. “If you look at what's going on in the United States right now, we have a record number of robots being ordered, and unemployment has fallen.”

Researchers and industry representatives say automation's potential is so great, and competition abroad so fierce, that the United States needs to invest even more in research and development.

“Investment in Europe, China and … South Korea is considerably larger [than in the United States], and it's going to get larger still,” warns Martial Hebert, director of Carnegie Mellon University's Robotics Institute.

The sharply divergent views — robots are a threat to the economy; robots are saviors — present policymakers and the country as a whole with conflicting challenges: On one hand, if the federal government does not increase support for basic research, automation proponents say, American robotics companies and the national economy risk falling behind Asian and European countries. Yet, say many analysts, policymakers must plan for the disruptions to some economic sectors and the job losses that automation may bring.

Despite the risks, Omohundro says the United States can't afford to hold back on research and development. “If any country decides not to use this technology, they're just going to fall behind and become less efficient,” he says.

Industrial Robot Sales Soar  

At the same time, he continues, the United States needs to prepare for the likely consequences. “The challenge for any country is dealing with the people who can't do their jobs better than a robot,” he says. “How do we economically support that population as this transition is happening?”

No one disputes that the robotics field is making tremendous strides: Robots — machines that can navigate and manipulate objects — and smart machines — computers that can analyze data and make choices based on that analysis — are performing tasks that few people dreamed they could do as recently as 10 years ago.

While the manufacturing sector has employed industrial robots for decades, smart machines are getting smarter and more numerous. Indeed, the lines between robots and smart machines are dissolving. Researchers are investing robots with more intelligence and computers with greater mobility.

Smart machines are increasingly working in public — both on the streets and in the workplace.

Google is testing its self-driving cars without human backup drivers in Austin, Texas. Its self-driving cars have already negotiated more than 1 million miles of city streets, without any mishaps attributed to the technology.5

The testing, says Adam Thierer, a senior research fellow with the Technology Policy Program at the Mercatus Center at George Mason University in Fairfax, Va., is just the beginning of a development that is going to have significant economic effects. “I can easily see the day within the next five years maybe that we have on the Interstate Highway System autonomous or semiautonomous vehicles that are traveling at specific hours of night in specific lanes to deliver long-haul freight,” he says. “Truckers will be disrupted before everybody else in the driverless car and driverless truck world.”

Equally interesting — though it has attracted relatively little media attention — is WorkFusion, an automated project manager developed at the Computer Science and Artificial Intelligence Lab at the Massachusetts Institute of Technology (MIT). WorkFusion selectively assigns work — such as writing data reports or scanning social media sites for specific content — to humans. It then evaluates the research and writing and reassigns it if it is not up to standards.6

“As the workers complete their assigned tasks, WorkFusion's machine-learning algorithms continuously look for opportunities to further automate the process,” Ford wrote in Rise of the Robots. “In other words, even as the freelancers work under the direction of the system, they are simultaneously generating the training data that will gradually lead to their replacement with full automation.”7

Just as WorkFusion is performing work traditionally done by data analysts and writers, Lionbridge's GeoFluent system can handle foreign language translations. The cloud-based service instantly translates speech or written text and can even translate passages into multiple languages for work groups that don't share a language.

Machines are increasingly handling more prosaic jobs, too. Thanks to more powerful sensors, smarter software and more versatile robotic limbs, joints and digits, machines are increasingly performing jobs such as emptying trash, cooking food or working on assembly lines.

Sales of industrial robots, which roughly tripled between 1995 and 2013, reached 229,000 units in 2014. The automotive and electronics industries, especially in China and South Korea, accounted for most of the increase, according to the International Federation of Robotics (IFR), an industry trade group. By the end of 2013, IFR said, the total worldwide workforce of industrial robots was between 1.3 million and 1.6 million units.8

South Korea Leads in Industrial Robots  

South Korea, the most roboticized country, has 437 industrial robots per 10,000 human workers, according to IFR. Japan is second, with 323 robots per 10,000 workers. The United States ranks seventh, with 152 robots per 10,000 workers.9

As experts debate the consequences of robotics for employment and society, many agree that some sectors will face bigger challenges than others. Among the jobs most at risk in the near future, according to one study, are telemarketers, tax preparers and office clerks.10

“We are in a paradigm shift,” says James Hughes, executive director of the Institute for Ethics and Emerging Technologies, a think tank in Willington, Conn. “If we begin to slowly displace workers and then more rapidly displace workers because of a variety of technologies — not just automation but also 3-D printing, for example — we need to have an affordable economic [model] for how people are going to equitably share in the benefits of that automation process and the downsides of it.”

Among the solutions Hughes and others suggest are retraining programs, shortening the workweek and work year and providing more paid vacations.

“Then we need to rethink the relationship between work, income, pensions, retirement and Social Security,” says Hughes. “One of the things that we have been promoting is the idea of an income guarantee — to gather all the various forms of redistribution into one basic redistribution check that everyone would get, and that this would have a kind of catalytic effect on unlocking people's creativity.”

Hughes acknowledges that such proposals would encounter significant resistance. “A lot of people in public policy, and people in general, don't want to embrace this,” he says.

As robotics enthusiasts, researchers, ethicists and others debate automation's impact on the economy, here are some of the questions being asked:

Is there a limit to what machines can do?

The popular TV game show “Jeopardy” inspired a spirited public debate over just how intelligent computers can be when on Feb. 16, 2011, IBM's Watson supercomputer thrashed the top-two all-time “Jeopardy” champions in the final edition of a special three-episode competition. In fact, Watson earned three times as much money as its human competitors.

The most striking aspect of Watson's performance was its ability to handle questions that involved double entendres and subtle tricks in phrasing.

The debate over “real” intelligence, consciousness and the limits, if any, of machine intelligence holds great implications for the economy: If machines become autonomous and capable of doing virtually every job humans do — and in many areas they can do more — will there be any work for humans?

“There are some AI [artificial intelligence] scientists who say that these systems will never achieve the same level of intelligence as humans — that there is some intrinsic barrier,” says Omohundro of Self-Aware Systems. “Most scientists say that we'll get there someday, but maybe it's 100 years away.”

Omohundro is reluctant to predict when machines might achieve human levels of intelligence. When pressed, however, he says it could happen as early as five years — or as long as 100 years.

Nate Soares, executive director of the Berkeley, Calif.-based Machine Intelligence Research Institute, which seeks to ensure that intelligent machines are beneficial, puts the date closer to 100. “It seems like we've got the hardware to do it, but we don't know how to write the program,” he says. “And it's much harder to predict insights than it is hardware developments. If I had to bet, I'd bet it happens in about 70 years.”

English clinical neuroscientist and philosopher Raymond Tallis doubts machines will ever attain anything approximating human intelligence. “Some have argued that thought does not require consciousness, so that computers can think, or will one day think, even though they will never be conscious,” he writes.

“It requires no equipment or subtle argument to demonstrate that this is nonsense. All you need is to focus on the thoughts you are having now. To deny that thought is conscious is self-refuting: you cannot deny the consciousness of your thoughts without being conscious of doing so.” As a result, Tallis said, an unconscious computer can't think. “It may assist us to think but is not itself thoughtful.”11

Without the ability to think, robots will be less able to perform jobs requiring critical thinking, such as working as a nuclear scientist or a graphics designer, some researchers say. Moreover, they add, humans will be the ones doing the hardest task: designing, building and servicing robots and smart machines. Karl Fogel, a partner at technology consultants Open Tech Strategies, told the Pew Research Center, “The reason [businesses] are investing in machine agents is precisely that they will replace more [lower-paid] humans than the number of [more highly paid] humans needed to build and maintain the machines.”12

Omohundro says robots may lack consciousness and be poor thinkers, “but there is no question that [machines] will become intelligent,” giving them the ability to handle many jobs.

The “Internet of Things” — a cloud-based network of connected sensors that connect to the Internet — is allowing robots to act independently of humans and to handle more complex jobs. This interconnectivity enables robots to share information with other robots and to even collaborate on work, two skills that improve robots' ability to master complex tasks. One such cloud-based platform is RoboEarth, a robot-specific Internet database and network sponsored by the European Union.13

IBM's Watson supercomputer handily beat the top-two all-time (Getty Images/Los Angeles Times/Carolyn Cole)  
IBM's Watson supercomputer handily beat the top-two all-time “Jeopardy” champions in 2011. In 2014, IBM announced that some 100 clients were planning to use the computer, including Memorial Sloan Kettering Cancer Center in New York City. Last July CVS said it will use Watson to monitor possible risks involving customer prescriptions. (Getty Images/Los Angeles Times/Carolyn Cole)

AI researchers rarely talk about consciousness. If a machine acts intelligently, it makes little sense to question whether it is intelligent, they say.

“The concerns we have are more about how quickly you could go from human-level intelligence to much-greater-than-human-level intelligence,” says Soares. And that, he says, could present dangers, especially if programmers design machines that can learn and modify their own programming.

The ability to think independently, say some experts, could turn robots into legal and economic liabilities.

“If you are going to gain really high confidence in something that is going to modify itself, you need to have a really, really good reason to think that when it self-modifies it is going to modify into something safe [that is, trustworthy and reliable], because five or 10 self-modifications down the line you're going to have no idea what it's doing,” Soares says.

Omohundro agrees. “You have to choose the goals for those systems very, very carefully,” he says. “Even very innocuous-sounding goals can lead to unintended consequences. And if you have a system that is goal-directed and not very carefully defined, it will have these drives toward self-protection, resource acquisition, replication.”

Omohundro and others note that little research has been done on how to ensure smart machines of the future remain reliable and pose no risks to humans. “There is a fair amount of funding for developing these technologies but hardly any for making sure that it is safe and that it has a positive beneficial impact,” he says.

Will machines disrupt the U.S. economy?

Economists and AI researchers agree that intelligent machines offer great potential to increase productivity in an array of industries and to relieve humans of highly dangerous tasks as well as tedious jobs, but they disagree on whether unemployment will rise as a result of robots.

Robots already have been deployed to inspect hazardous environments, such as nuclear facilities, and researchers are developing robots to perform search-and-rescue operations in disaster zones. And, of course, robots perform many repetitive tasks in manufacturing plants, such as assembling cars.

Not only do machines relieve humans of such chores, they are generally able to perform the work more efficiently and at lower cost, making the business more competitive in global markets and making products more affordable for consumers.

Robotics industry representatives believe automation is one key to a stronger economy. Among six of the countries with the highest adoption rates of robotics — Brazil, China, Germany, Japan, South Korea and the United States — overall employment rose between 2000 and 2012 in all but Japan, a 2013 International Federation of Robotics (IFR) report said.14 Several factors fueled this increase, according to the IFR, including rising demand for products from population increases, the creation of new technologies, and the efficiencies gained from automation, which have helped manufacturers raise output. The gain in employment was large enough, the IFR said, to offset drops in manufacturing employment in two of those countries: Germany and the United States.

In addition to helping countries' economies, the robotics industry employs 170,000 to 190,000 people worldwide, according to the IFR 2013 report.15

Regardless of their impact on employment, “robots are increasingly essential to the competitiveness of a country's manufacturing sector,” according to Scott Andes and Mark Muro, researchers at the Brookings Institution think tank in Washington. “The fact that countries like Germany, Sweden, and Korea are deploying automation technology at a much faster rate than the United States points to serious competitive challenges — and further debate about the use and impacts of automation.”16

Lawrence Mishel, president of the Economic Policy Institute, a think tank in Washington with board members from several major labor unions, predicts that as automation spreads, it will create more jobs and make products more affordable. “I've not seen a case for why this type of technology will not lower costs and provide people more spending power that will create other jobs so we won't see massive technological unemployment,” he says.

However, other analysts caution that establishing the causes of employment trends is tricky, with many factors affecting the jobless rate. “Correlation is not causation, and there is no shortage of alternative explanations,” Andes and Muro wrote. Still, they argued, “the evidence suggests there is essentially no relationship between the change in manufacturing employment and robot use” between 1993 and 2007.17

Mishel concurs: “There is no prima facie evidence that automation or robots has in any way been responsible for higher unemployment or that they are associated with the rising inequality of wages in the 2000s.”

He predicts that as automation spreads, it will create more jobs and make products more affordable. “I've not seen a case for why this type of technology will not lower costs and provide people more spending power that will create other jobs so we won't see massive technological unemployment,” he says.

But a growing number of analysts say robots and intelligent machines have the potential to alter broad swaths of the economy in ways that previous game-changing technologies have not.

“It's possible that this time it is different,” warns Erik Brynjolfsson, director of the MIT Initiative on the Digital Economy, pointing to the potential application of intelligent machines across many job sectors. “The nature of the technology is different than in the past. I am hopeful that entrepreneurs will help invent new jobs, new goods and services, but will they employ more and more people? It's not an automatic event.”

Author and entrepreneur Ford made much the same point. “In the past, automation technologies tended to be relatively specialized and to disrupt one employment sector at a time, with workers then switching to a new emerging industry,” he wrote. “The situation today is quite different. Information technology is a truly general-purpose technology, and its impact will occur across the board.”18

Marc Andreessen, a software engineer and prominent tech entrepreneur, isn't buying it. The idea that computers and robots are going to “eat all the jobs,” he said, is based on a “lump-of-labor” fallacy — the notion that the number of jobs is finite. Andreessen argued that human wants and needs are infinite, and that machines can't fill all of them. “Even when robots and AI are far more powerful, there will still be many things that people can do that robots and AI can't,” Andreessen wrote in 2014. “For example: creativity, innovation, exploration, art, science, entertainment and caring for others. We have no idea how to make machines do these.”19

Brynjolfsson and Andrew McAfee, coauthors of The Second Machine Age: Work, Progress, and Prosperity in a Time of Brilliant Technologies, wrote that machines are unlikely to eliminate all human workers, at least for the next decade. And certain types of jobs — those involving interpersonal or social skills, such as substance abuse counselors or recreational therapists — will require humans long into the future.20

“But will there be enough demand, especially over the long term, for those two types of human labor: that which must be done by people and that which can't yet be done by machines?” they write. “There is a real possibility that the answer is no — that human labor will, in aggregate, decline in relevance because of technological progress, just as horse labor did earlier. If that happens, it will raise the specter that the world may not be able to maintain the industrial era's remarkable trajectory of steadily rising employment prospects and wages for a growing population.”21

Dean Baker, co-director of the Center for Economic and Policy Research, a liberal think tank in Washington, says the primary difference between techno-optimists and those who predict severe economic and social disruption is their views on how quickly advances in robotics and machine intelligence will occur.

“I don't see technology as presenting qualitatively new challenges,” Baker says. “The question is, what is the rate at which [new technologies] will come online?” A technology surge could result in higher unemployment, he says, but it is not happening now. “And it can't be something that happens overnight. We have a long way to go.”

Automation Threatens Low-Skill Jobs  

However, two University of Oxford researchers recently published a study estimating that 47 percent of the current jobs in the United States are at risk of being automated over the next 10 to 20 years, with the jobs most immediately in danger being those that have few technical requirements, such as telemarketing and office clerical work. Robots are far less likely to replace humans in specialized roles such as nuclear engineer (7 percent probability), pharmacist (1.2 percent) and dentist (0.4 percent), according to their study.22

Should the government fund a robotics revolution?

The primary federal effort to boost research and development of non-military robotics and machine intelligence is the National Robotics Initiative, begun in 2011 by President Obama and administered by the National Science Foundation. The program awarded $31.5 million in grants in 2014, and officials expect to issue up to $50 million in 2015.23

“Revolutions in technology have transformed the way we live and the way we work,” Obama noted in announcing the National Robotics Initiative at a speech in Pittsburgh in 2011. “Businesses and industries can relocate anywhere in the world, anywhere that there are skilled workers, anywhere that there is an Internet connection. And companies have learned to become more efficient with fewer employees.”24

Developing a strong robotics industry, Obama said, is “about making sure our workers and businesses have the skills and the tools they need to compete better, faster and smarter than anybody else.”

Not surprisingly, researchers say they need more government funding.

“Certainly, it is not enough,” says the Robotics Institute's Hebert, noting that government funding for robotics research has been flat or falling in recent years. That fact may surprise some, Hebert says: “Many large companies like Amazon and Google … have been investing considerable amounts, so there is a temptation to say that the field is becoming more mature and that therefore the private sector is going to take over,” he says. “That's not true at all, at this point at least.”

Researchers used a 3-D printer (Getty Images/The Christian Science Monitor/Ann Hermes)  
Researchers used a 3-D printer, along with basic electronic components, to make a robot at the Massachusetts Institute of Technology's Computer Science and Artificial Intelligence Laboratory. “The broad adoption of robots will require a natural integration of intelligent machines into the human world rather than an integration of humans into the machines' world,” wrote lab director Daniela Rus. (Getty Images/The Christian Science Monitor/Ann Hermes)

“The explosion of robotics in the private sector,” he adds, “comes from basic research results attained through government funding. And we do still need that because we have to get to the next level now.”

Although he says government funding isn't enough, Hebert calls the National Robotics Initiative a good start. So does Jeffrey Krichmar, a professor of cognitive sciences at the University of California, Irvine. “It is a big program,” he says. “We all wish it was bigger.”

Noting that South Korea and other Asian countries are far ahead of the United States in robotics, Krichmar suggests that part of the problem may be that military research gets the lion's share of funding. “In this country, most of the major robotics funding is through DARPA [Defense Advanced Research Projects Agency] or other military,” he says. “It would be nice if it was spread more.”

Indeed, governments in Asia — including Japan, South Korea and China — as well as the European Union have invested in public-private robotics companies. These governments “each recognize robotics as a transformative technology, a kind of emerging infrastructure,” says Ryan Calo, an assistant law professor at the University of Washington. “Accordingly, they have coordinated massive public-private partnerships in robotics and begun formally to develop legal and policy frameworks.” In addition, many governments offer tax incentives and subsidies to robotics companies.

If U.S. research on non-military robotics is underfunded, Soares at the Machine Intelligence Research Institute says that research funds for ensuring the safety of robots and intelligent machines are even more scarce. According to Soares, the first program in the field was funded only last June, and that was financed by PayPal co-founder Elon Musk, founder of Tesla Motors and SpaceX, and the Open Philanthropy Project, with no help from the government. “The safety research is vastly outstripped by the capabilities research,” he says.

At the same time, Soares says, the robotics field is still so small that it almost certainly couldn't effectively use, say, $1 billion in research funding. “I think if you suddenly dumped $1 billion on this field, you would get a lot of vultures coming in,” he says. “But we could probably soak up $100 million.”

Some in the robotics industry also want the government to provide incentives to businesses to adopt robotics.

“Of course, we'd like to see more in terms of government support for robotics,” says Burnstein of the Association for Advancing Automation. “But maybe more importantly, we would like to see user incentives to implement robotics. If we want to stay competitive and be a manufacturing leader globally, we're going to have to automate.”

Burnstein says the Congressional Robotics Caucus, founded by Rep. Mike Doyle, D-Pa., isn't very active. “It would certainly be good if more representatives of Congress were involved in that,” he says.

Doyle declined requests for comment, but the caucus' website says the caucus seeks to educate members of Congress about robotics and to work toward ensuring that the United States remains competitive in the field.25

The Institute for Ethics and Emerging Technologies' Hughes says he has had difficulty getting politicians to discuss unemployment caused by automation and how to alleviate it. “I've been going around to politicians here in Connecticut, and we've also had a national effort to try to find folks in D.C. to talk about technological unemployment,” he says. “It is a third rail at this point, in terms of people not wanting to give any credence to the possibility.”

Go to top

Background

Eary Fears

Even before Scottish inventor James Watt patented an improved steam engine in 1781, inventors strove to create machines that could work more efficiently than humans — causing some to worry that those machines would threaten workers' livelihoods. In 1589, English clergyman and inventor William Lee traveled to London to seek patent protection for his mechanical knitting machine. Queen Elizabeth, however, declined to grant him a patent, saying, “Consider thou what the invention could do to my poor subjects. It would assuredly bring to them ruin by depriving them of employment, thus making them beggars.”26

The introduction of steam-powered machinery launched the Industrial Revolution, and British manufacturing boomed, turning Great Britain into the world's leading economic superpower almost overnight. The new technologies also transformed the nature of work — not only centralizing work in factories instead of in dispersed cottages, but also fractionalizing work so that individual workers often created only a small part of a product — the effects of which rippled through British society. Within two decades, the technologies spread to other countries in Western Europe and North America, as Britain's highly guarded engineering secrets eventually leaked out.

As Carl Benedikt Frey and Michael A. Osborne, engineering professors at the University of Oxford, note, the technologies of the Industrial Revolution were “deskilling.” “Work that had previously been performed by artisans was now decomposed into smaller, highly specialised sequences, requiring less skill, but more workers, to perform,” they wrote.27

Thus, while steam-powered looms did not reduce overall employment, they lowered the need for highly skilled workers in the textile industry. The change led to social unrest, including a wave of machine-breaking protests that erupted in the English Midlands from 1811 through 1816. The Luddites — rebels named after Ned Ludlum, a machine-breaker from Leicestershire — destroyed looms in nighttime raids on textile mills. The violence was put down by British troops, but not without bloodshed.

“In the end, the fears of the Luddites that machinery would impoverish workers were not realized, and the main reason is well understood,” said economists Joel Mokyr, Chris Vickers and Nicolas L. Ziebarth. “The mechanization of the early 19th century could only replace a limited number of human activities. At the same time, technological change increased the demand for other types of labor that were complementary to the capital goods embodied in the new technologies.”28

The new tools created during the Industrial Revolution also created new industries. Mass production of metal parts enabled rapid advances in manufacturing, transportation and communications, among other sectors.

Rapid growth, however, did not quiet the fears of workers and some economists.

Among others, Karl Marx predicted that the labor-saving aspect of machinery — “automation” was not yet a term in use — would lead to unemployment. It would also lead, he warned, to downward pressure on wages and increased employment of unskilled women and children.

“In so far as machinery dispenses with muscular power, it becomes a means of employing labourers of slight muscular strength, and those whose bodily development is incomplete, but whose limbs are all the more supple,” Marx wrote. “The labour of women and children was, therefore, the first thing sought for by capitalists who used machinery.”29

Historians describe the period just after World War I (1914-18) as the height of the Machine Age — an era of rapid industrialization in Western Europe and North America. The Great Depression that began in 1929 temporarily halted that expansion and gave rise to renewed worries about machinery. As New York Times reporter Nicholas Carr related, “The mayor of Palo Alto, California, wrote a letter to President Herbert Hoover imploring him to take action against the ‘Frankenstein monster’ of industrial technology, a scourge that was ‘devouring our civilization.’”30

Computers and Industrial Robots

The challenges of mass-producing vehicles, weapons and other products for the military in World War II spurred countless improvements in machinery and industrial processes. And government support for research and development in an array of fields — but especially electronics — laid the groundwork for an explosion of new capabilities that would be commercialized in the 1950s.

In fact, the military financed the first general purpose electronic computer — a room-sized device named ENIAC (Electronic Numerical Integrator And Computer), created in 1946 to calculate artillery firing tables.

In 1948, MIT mathematician Norbert Wiener published Cybernetics: or Control and Communication in the Animal and the Machine, which laid out the theoretical structure for creating self-regulating systems, both mechanical and electronic. Apparently concerned about the potential of the emerging technologies, and about humans' abilities to manage them, he wrote an article for The New York Times — only recently published — headlined “The Machine Age.”

A robotic arm prepares to lift a patient from her bed to a wheelchair (Getty Images/Bloomberg/Kiyoshi Ota)  
A robotic arm prepares to lift a patient from her bed to a wheelchair at a nursing home in Chiba, Japan. Japan is the second-most roboticized nation in the world, after South Korea; the United States is a distant seventh. Robotics industry representatives say automation is a key to saving jobs, though critics warn that robots may eliminate jobs. (Getty Images/Bloomberg/Kiyoshi Ota)

“If we move in the direction of making machines which learn and whose behavior is modified by experience, we must face the fact that every degree of independence we give the machine is a degree of possible defiance of our wishes,” Wiener warned. “The genie in the bottle will not willingly go back in the bottle, nor have we any reason to expect them to be well disposed to us. In short, it is only a humanity which is capable of awe, which will also be capable of controlling the new potentials which we are opening for ourselves. We can be humble and live a good life with the aid of the machines, or we can be arrogant and die.”31

Wiener's warning came at a time when primitive electronic computers were just emerging and even rudimentary robots did not exist. Yet he foresaw a world filled with “machines much more closely analogous to the human organism.”

“They will control entire industrial processes and will even make possible the factory substantially without employees,” he said. Indeed, Wiener fretted that unless the economic system was changed, “we are in for an industrial revolution of unmitigated cruelty.”32

Wiener, who died in 1964, lived long enough to see the first commercial use of an industrial robot.

In 1956, George Devol and Joseph Engelberger founded the first robotics company, Unimation, short for “universal animation.” Five years later, General Motors became the first company to use Unimation's robot, Unimate, the world's first industrial robot. It moved die castings from an assembly line and welded them onto auto bodies. Unimate consisted of a 4,000-pound arm controlled by commands stored on a magnetic drum.

Also in 1956, a conference at Dartmouth College to discuss artificial intelligence is considered AI's founding event. The primary attendees and organizers — John McCarthy, Marvin Minsky, Allen Newell and Herbert Simon — became leaders in the field for the next several decades. But what really made the conference remarkable was the group's high expectations for AI.

“The study is to proceed on the basis of the conjecture that every aspect of learning or any other feature of intelligence can in principle be so precisely described that a machine can be made to simulate it,” reads the conference proposal. “An attempt will be made to find how to make machines use language, form abstractions and concepts, solve kinds of problems now reserved for humans and improve themselves. We think that a significant advance can be made in one or more of these problems if a carefully selected group of scientists work on it together for a summer.”33

It took much longer than a summer to make even a dent in the complex task of creating intelligent machines. In fact, it wasn't until 1966 that the first robot with machine intelligence was created in the form of “Shakey,” a mobile machine created by researchers at the Stanford Research Institute in Menlo Park, Calif. The robot, so named because of its jerky locomotion, was a simple platform equipped with a TV camera, a laser range finder and sensors to detect impacts with objects in the environment. An onboard program processed the collected data and allowed Shakey to navigate as it moved.

Despite the optimism at the Dartmouth conference, and a good deal of funding provided by the Department of Defense, researchers found the obstacles to making machines intelligent more difficult than expected. The field entered a period the researchers refer to as the “AI winter.”

Nevertheless, the period did include the creation of the first full-scale humanoid robot — Wabot-1. Created by Ichiro Kato and his colleagues at Waseda University in Japan in 1973, Wabot-1 moved on two legs and could grip and carry objects. It could also read music and play an electronic piano.

Expanding Research

According to Nick Bostrom, director of the Programme on the Impacts of Future Technology at the University of Oxford in England, the Japanese pulled the field out of its AI winter. “A new springtime arrived in the early 1980s, when Japan launched its Fifth-Generation Computer Systems project, a well-funded public-private partnership that aimed to leapfrog the state of the art by developing a massively parallel computing architecture that would serve as a platform for artificial intelligence,” he writes.34

Google's self-driving cars have negotiated more than 1 million miles of city streets (Getty Images/Justin Sullivan)  
Google's self-driving cars have negotiated more than 1 million miles of city streets, including in Austin, Texas, without any mishaps attributed to the technology. In 2011 Nevada became the first state to allow self-driving cars to operate on public streets. Some researchers say that within the next five years autonomous or semiautonomous vehicles could be hauling freight on the Interstate Highway System, using specific lanes during designated nighttime hours. (Getty Images/Justin Sullivan)

Research on robotics and AI also began to expand in the United States in the early 1980s, thanks to improvements in computing power and encouraging advances in software development, particularly software that emulates the human decision-making process.

Funding from DARPA began to pour in to research labs, especially as the United States became involved in long-term military operations in Afghanistan and Iraq.

Peter W. Singer, a political scientist and foreign policy analyst then at the Brookings Institution, said the “magic moment” for the development of military robotics occurred in 1995, when unmanned aircraft were integrated with GPS.35 That same year, the two workhorses of the government's drone fleets appeared — the Predator and the Global Hawk.

DARPA funding spurred the development of ground-based robots as well, initially for reconnaissance and the detection and disposal of explosive devices.

DARPA organized and funded “grand challenge” competitions to develop robotics technologies, with the winners receiving prize money for further development. The early grand challenges focused on self-driving cars and were held in the Mojave Desert.

No award was given in 2004, the first year of the competition, because none of the vehicles managed to complete the course. The second challenge, in 2005, was won by a Stanford University team led by Sebastian Thrun, director of the Artificial Intelligence Laboratory at Stanford and a founder of the Google X Lab research facility. Each successive grand challenge increases the difficulties of the tasks to be performed. A team from Carnegie Mellon University won the 2007 Urban Challenge, which simulated an urban environment for competing vehicles.

More recently, DARPA launched a new grand challenge, aimed at developing robots capable of assisting humans in responding to natural and man-made disasters. In 2013, the challenge was to create software simulating a humanoid disaster response. A team from South Korea won; the Institute for Human and Machine Cognition in Pensacola, Fla., came in second.

The winner of the first competition involving actual robots, in 2013, was SCHAFT, a robot built by a Japanese team and bought by Google. The competition asked robots to drive a car, climb a ladder, remove debris and open a door, as well as four other tasks. First place in the 2015 trials — which required teams to develop software to control the Atlas robot developed by Boston Dynamics — went to a team from South Korea. Winners in each competition earned $2 million. For the 2015 competition, each team had to show that its robots could manipulate an emergency shut-off switch, walk 10 meters without stumbling, turn a valve 360 degrees and climb over a barrier.

Go to top

Current Situation

Playing Catch-Up

While the United States may be falling behind Asia and Europe in public investments in robotics research and development, U.S. robotics companies are growing rapidly and employing increasing numbers of workers.

ARK, a technology investment consulting firm, estimated in 2014 that the number of robotics-related jobs in the United States had risen to 26,000 and said robotics companies, “should be in an accelerated hiring mode in the next few years.”36 What's more, the company projects that the U.S. robotics industry will employ nearly 2 million workers by 2025, and nearly 5 million by 2035.

U.S. tech companies also have been on a buying spree, acquiring both domestic and foreign robotics manufacturers. Google, for example, in 2013 acquired eight robotics companies, including Boston Dynamics and the Japanese start-up that created the award-winning SCHAFT robot.37 And in 2012 Amazon paid $775 million to acquire Kiva Systems, a Massachusetts-based robotics company.38

A robot plays a violin in Tokyo (AFP/Getty Images/Kazuhiro Nogi)  
A robot plays a violin in Tokyo, demonstrating the increasing dexterity of robotic hardware and software. Thanks to more powerful sensors, smarter software and better-engineered robotic limbs, joints and digits, machines are increasingly performing jobs that have required human arms and hands, such as emptying trash, cooking food or working on assembly lines. (AFP/Getty Images/Kazuhiro Nogi)

Meanwhile, researchers continue to develop artificial intelligence, robotics and advanced processing hardware. In AI, IBM has recently begun touting Watson's skills in a large marketing campaign, having devoted a reported $1 billion and 2,000 employees to push sales of the powerful computer.39 Watson employs natural-language processing algorithms and neural networks — software simulations of the human nervous system — to analyze data.

In October 2014, IBM opened a business office for Watson in downtown Manhattan and announced that it had some 100 clients planning to use the computer. The health care industry is particularly interested in Watson's capabilities, including the Memorial Sloan Kettering Cancer Center in New York City, which participated in a trial demonstration using Watson to recommend treatments. The center is reportedly planning to extend the trial.40

GenieMD, an app that offers medical advice based on a user's medical history and current medications, has signed on as a Watson user, and the CVS drugstore chain announced in July 2015 that it will use Watson to monitor possible risks involving customer prescriptions.41 SparkCognition, another early client, offers big-data analytic tools to monitor critical infrastructure for vulnerabilities and to protect computer networks.

While Watson is the biggest name among natural-language analytic engines, competition is beginning to emerge. In 2014, Google acquired London-based DeepMind, an artificial intelligence company, for more than $500 million, but Google has not said how it intends to deploy DeepMind.42

Some economists predict that workers whose jobs involve problem-solving and human-level manipulation skills — such as plumbers, electricians and automotive technicians — will be among those most secure from automation.43 However, by integrating AI and engineering, robotics researchers are rapidly improving robots' ability to move and manipulate objects, as well as to cooperate with other robots.

In August, an MIT team introduced software enabling robotic grippers to adjust to changing task conditions. If, say, the gripper picks up a pencil at the end rather than in the middle, the software can “tell” the grip to loosen and push the pencil against an object to reposition it. The team is working on ways to allow robots to adjust their interactions with objects to account for other environmental factors, such as gravity and surface friction, so they can more easily control tossed or sliding objects.44

Eatsa, a fully automated restaurant in San Francisco's Financial District (Getty Images/Justin Sullivan)  
Eatsa, a fully automated restaurant in San Francisco's Financial District, has no servers, wait staff or cashiers. Food is delivered at the push of a button to a customer's personalized compartment. In the future, say robotics scientists, two of the biggest challenges will be getting robots to communicate better with humans and improving their ability to overcome unexpected events. (Getty Images/Justin Sullivan)

Another MIT team is developing software — called 3D SLAM, or simultaneous location and mapping — to improve robots' recognition of objects and locations. SLAM systems already exist, but they create their object maps from live camera data and by making depth measurements — intensive and time-consuming tasks. The MIT team's method employs machine learning to train robots to recognize objects by processing large libraries of single-perspective images captured by standard cameras, a process that needs to be accomplished only once.45

Experts say the emerging field of cloud robotics also offers great potential. Software that enables robots to perform certain actions — whether to manipulate an object, navigate terrain, process data or recognize objects — is uploaded to the cloud, where other robots can access it. As a result, lower-cost robots can take advantage of the collective experience of other robots.

In addition, Internet-connected robots can use the cloud as a platform to coordinate their activities with each other or to offload heavy-duty processing tasks.

RoboEarth, currently the largest cloud robotics effort, began in 2009 after receiving a four-year, $7.6 million grant from the European Union. RoboEarth already has produced an array of tools and resources for roboticists, and at the top of that list is Rapyuta, RoboEarth's cloud engine, an open-source platform that allows robots to offload heavy computation tasks.

Another cloud-based spinoff from RoboEarth is RoboHow, a four-year project — also funded by the European Union — that got underway in 2012. RoboHow is aimed at enabling robots to perform everyday human tasks, such as cooking.

And in 2014, Ashutosh Saxena, an assistant professor of computer science at Stanford University, along with colleagues at Cornell, Brown and the University of California, Berkeley, launched RoboBrain, a project to create a cloud repository of data for robots. The site includes approximately 1 billion images, 120,000 YouTube videos and 100 million how-to documents and manuals.

Need for Speed

Writing the software that empowers problem-solving neural networks that can allow a robotic hand to catch a baseball is a major challenge, and researchers acknowledge that machines need faster hardware to do the computing.

Help is on the way. In August 2014, IBM announced it was only two to three years away from making its TrueNorth processor — the chip-based processing unit at the heart of computers — commercially available.46

TrueNorth works somewhat like the human brain and carries 5.4 billion transistors, 1 million programmable “neurons” and 256 programmable “synapses” that mimic the connections between neurons in the brain. That's more than a little shy of the 100 billion neurons in the human brain, but it's still far beyond anything that has been available to robot designers to date. And the design is not just about the numbers of transistors and synapses and the processing power they provide. The processor's architecture supports parallel processing, which allows software running on it to resolve ambiguities and solve problems in human-like ways.

TrueNorth also allows machines to distinguish between several voices and accurately recognize speech without first having to train with individual voices, according to IBM.

What's more, the TrueNorth processor runs on only 70 milliwatts of power, roughly the same as a hearing aid. That makes it ideal for use in even small robots, such as search-and-rescue devices.

TrueNorth is the first artificial “brain” developed by the SyNapse program, begun by the Defense Advanced Research Agency (DARPA) in 2008 with help from IBM, Hewlett-Packard and HRL Laboratories. DARPA had devoted $102.6 million in grants to SyNapse by January 2013, according to a website that tracks research on the human brain.47

Looming Limitations

Despite the remarkable advances occurring in robotics labs, researchers acknowledge that they still have to overcome major hurdles before robots become pervasive and capable of interacting in human environments.

“The broad adoption of robots will require a natural integration of intelligent machines into the human world rather than an integration of humans into the machines' world,” wrote Daniela Rus, director of the Computer Science and Artificial Intelligence Laboratory at MIT. “Despite recent significant progress toward that goal, problems remain in three important areas. It still takes too much time to make new robots, today's robots are still quite limited in their ability to perceive and reason about their surroundings, and robotic communication is still quite brittle.”48

Rus and a team of researchers from Harvard, MIT and the University of Pennsylvania are working to create a “robot compiler” designed to speed development of robots.

That partially involves two related challenges — getting robots to communicate better with humans and improving their ability to overcome the unexpected. “If a robot encounters circumstances that it has not been programmed to handle or that fall outside the scope of its capabilities, it enters an ‘error’ state and stops operating,” Rus said. “Often, the robot cannot communicate the cause of the error. Robots need to learn how to adjust their programs so as to adapt to their surroundings and interact more easily with people, their environments, and other machines.”49

Go to top

Outlook

“Here to Stay”

Robot makers and financial analysts are bullish about the prospects for growth in robot sales. According to a recent forecast by Research and Markets, an Irish market research company, revenues for global industrial robotics are expected to rise to $37 billion by 2018.50

BNY Mellon, an investment management firm, also predicts rapid growth in all sectors of robotics. “There is little doubt robots are here to stay and can only become more advanced,” wrote Jason Conway, senior investment editor at BNY Mellon. “While the investment potential of this sector is yet to be fully realized, it continues to present some fascinating prospects for future development and further technological progress ahead.”51

Researchers are likewise confident of rapid advances in the capabilities of intelligent machines in the near future, and even skeptics question only the timeframe of those advancements.

Many economists expect the new technologies to be introduced gradually and say existing economic tools available to policymakers are up to the task of handling any disruptions to the economy.

The Center for Economic and Policy Research's Baker says government doesn't need to step in at this point to deal with technological unemployment. “We have unemployment insurance,” he says. “We have had sectors where productivity rises rapidly and causes unemployment. Look at all the people who used to make film for cameras who aren't there anymore. I don't see this as being something qualitatively new.”

Thierer of George Mason University's Technology Policy Program agrees. “There's very legitimate reason to be concerned that there will be some professions and some jobs that will be disrupted by robotic innovation and potentially will disappear because of it,” he says. “The difference is, I'm not panicking about it in the sense that I don't necessarily think that something must be done right now to stop that.”

Thierer says that rather than trying to reconfigure the economy, and possibly inhibit innovation, policymakers should focus on training people for new jobs. “The better approach is to think constructively about what kind of new skill sets will make sense moving forward,” he says. “How will we repurpose various types of labor in different professions? That's clearly a more constructive approach.”

Others say, however, that if change is more rapid, existing policy options may be insufficient.

“The nature of recent technological change suggests that the adjustments that were possible in the past might not continue to take place,” wrote Stuart W. Elliott, a visiting analyst at the 34-nation Organisation for Economic Co-operation and Development, in Paris.52

A “broad range” of occupations could see job losses, Elliott said. And he added that while many economists look only at changes over the next decade, “several decades is relatively short for adapting to a change of such magnitude.”

Indeed, those who know the technologies best expect the changes to be coming sooner than many economists anticipate.

“It is reasonable to assume that robots will in the not-too-distant future be able to perform” memory tasks at human levels,” noted Gill A. Pratt, former program manager at DARPA. What's more, Pratt wrote that improvements in artificial intelligence are coming more quickly than expected. He added that the effects on the economy and human workers “are certain to be profound.”53

Software entrepreneur Ford sees big trouble ahead. “The frightening reality is that if we don't recognize and adapt to … advancing technology, we may face the prospect of a ‘perfect storm’ where the impacts from soaring inequality, technological unemployment and climate change unfold roughly in parallel, and in some ways amplifying and reinforcing each other,” he wrote.54

Because of these potential upheavals, Ford warns that Congress should be doing more to prepare for the potential for economic and social dislocations.

“Is anyone in Congress paying attention? I still don't see much evidence of that,” he says. “It's a pretty toxic issue that most politicians aren't going to want to touch.”

Go to top

Pro/Con

Should Congress establish a Federal Robotics Commission?

Pro

Ryan Calo
Assistant professor of law, University of Washington. Written for CQ Researcher, September 2015

Should we regulate robots? We already do, and not particularly wisely. Sometimes government officials threaten innovation by dragging their feet. American companies began to test drones abroad because the Federal Aviation Administration bans the technology for private use and has been slow to grant exceptions.

Other times, officials act in haste. The Food and Drug Administration approved robotic surgery as analogous to minimally invasive laparoscopic surgery, only to receive hundreds of reports of adverse events, such as sparks, loss of video feed and sudden, uncontrolled movements that are unique to robots.

The underlying problem — namely, that state and federal policymakers don't have a good handle on the technology — is hardly unique to the United States. What is unique is our failure to respond. The governments of Japan, South Korea and the European Union each recognize robotics as a transformative technology, a kind of emerging infrastructure. Accordingly, they have coordinated massive public-private partnerships in robotics and begun formally to develop legal and policy frameworks.

Meanwhile, in the United States, we continue to approach robotics policy in piecemeal fashion, relying on industry experts or single-issue activists to push for laws for a single jurisdiction or type of robot.

What we need in our policymakers is the expertise to make considered choices. Establishment of a Federal Robotics Commission is one way to accrue that expertise. Rather than pass rules for robots, which would be premature, the commission could advise federal and state lawmakers on a growing variety of issues, from driverless-car safety to air traffic control procedures for drones.

There are other ways for the government to gain traction on robotics. Each federal or state agency could hire a few roboticists. NASA, with its deep understanding of cyberphysical systems, or the White House Office of Science and Technology Policy could take on a greater role in robotics policy.

But a stand-alone agency is the most direct way to develop robotics expertise, in my view. An agency dedicated to robotics could also attract the best and the brightest. Government competes with industry and academia for technologists, but who wouldn't want to work at a Federal Robotics Commission?

If we do nothing, as some hope, robotics may wind up the first transformative technology since steam in which America is not at the forefront.

Con

Adam Thierer
Senior research fellow, Technology Policy Program, Mercatus Center, George Mason University. Written for CQ Researcher, September 2015

America didn't need a Federal Internet Commission to gain a commanding lead in the digital economy, and we don't need a Federal Robotics Commission to ensure U.S. firms become global leaders in robotics innovation, either.

In fact, a new Washington bureaucracy for robotics could be a counterproductive innovation-killer. There just isn't a good history of stodgy agencies being able to keep pace with quick-moving technologies like these. Instead, as America's experience with the Internet and computing proves, light-touch regulation is vastly superior to top-down, command-and-control regulation by unelected bureaucrats.

Moreover, if we do create a Federal Robotics Commission, there's a good chance it would come to regulate the Internet and computing anyway. Robotics and the many technologies and industries it already includes — such as driverless cars, commercial drones and the Internet of Things — are developing on top of the building blocks of the information revolution: microprocessors, wireless networks, sensors, “big data,” etc. We can't regulate one without regulating the other. “Regulatory creep” has proved to be a real problem in many other sectors where agencies come to take on a greater regulatory role over time.

Of course, the rapid growth of robotics and artificial intelligence might raise some serious policy issues that cannot be ignored, but that's still not a good reason to create bureaucracies to regulate them. Congress or the courts can handle those issues as they develop. To the extent robotic systems are involved in accidents that harm individuals or their property, product liability law can be utilized to resolve disputes. And the Federal Trade Commission and state attorneys general already possess general jurisdiction over “unfair and deceptive” practices in all other sectors of the economy. That same authority will cover the robotics sector as well.

If we need robotics experts to more closely study other issues pertaining to robotic ethics, plenty of other entities (including nongovernmental organizations) exist that could play that role, including the National Science Foundation.

But the United States isn't going to become a global innovation leader in robotics by erecting big bureaucracies that attempt to plan it all preemptively. Instead, we need a more fluid and decentralized approach that will let entrepreneurs experiment without having to rush to Washington to seek permission each time they come up with a great new idea.

Go to top


Chronology

 
1940s–1950sThe development of computers paves the way for robotics and artificial intelligence (AI).
1946U.S. Army creates ENIAC (Electronic Numerical Integrator And Computer), the first general purpose electronic computer, to calculate artillery firing tables.
1948MIT mathematician Norbert Wiener publishes Cybernetics: or Control and Communication in the Animal and the Machine, which lays out the theoretical structure for creating self-regulating systems, both mechanical and electronic. Wiener later warned in a newspaper article that machines may lead to “an industrial revolution of unmitigated cruelty” and result in large job losses.
1950British scientist Alan Turing publishes Computing Machinery and Intelligence, in which he asks the question, “Can machines think?”
1956George Devol and Joseph Engelberger establish the first robotics company, Unimation, short for “universal animation,” and develop the world's first industrial robot…. A conference at Dartmouth College, considered the founding event of the field of artificial intelligence (AI), sets the goal of designing a machine that can simulate “every aspect of learning or any other feature of intelligence.”
1960s–1970sRobotics suffers from an “AI winter,” a period when little is accomplished in the field.
1966SRI International demonstrates the first robot that employs artificial intelligence; named Shakey, it navigates by processing data collected by a TV camera, laser range finder and bump sensors.
1973Ichiro Kato creates the first full-scale humanoid robot — named Wabot 1 — at Waseda University in Japan. The walking robot could carry objects and play an electronic piano.
1980s–1990sAI winter ends as countries invest heavily in artificial intelligence.
1982Japan begins its Fifth Generation Computer Systems project, aimed at producing a parallel computer platform for artificial intelligence. Some credit the highly publicized project with spurring investments in other countries, including the United States, and helping end the AI winter.
1995The first unmanned Predator and Global Hawk drones are delivered to the U.S. military.
2000-PresentSmart machines — those capable of analyzing data and acting on it — head for commercial markets.
2004The Pentagon's Defense Advanced Research Projects Agency (DARPA) hosts its first grand challenge for self-driving cars, although none of the competitors is able to complete the course. The next year, a Stanford University team wins.
2009A four-year, $7.6 million grant from the European Union leads to the founding of RoboEarth, a cloud-based repository for robot algorithms and resources.
2011iRobot Corp. sends four robots to Japan to assist in responding to the devastation at the Fukushima nuclear plant, which was damaged in a March earthquake and tsunami. The robots inspected areas of the facility too radioactive for humans to enter…. Nevada becomes first state to allow self-driving cars to operate. The next year a Toyota Prius equipped with Google's self-driving technology receives the first license in the state…. WorkFusion, a platform created in MIT's Computer Science and Artificial Intelligence Lab, is commercialized, offering clients automated training and quality control of online workers.
2014An estimated 27,685 industrial robots valued at $1.6 billion are ordered from North American companies, an increase of 28 percent in units and 19 percent in value over the previous year.
2015In Austin, Texas, Google begins testing self-driving cars without human back-up drivers. Six states — California, Florida, Michigan, Nevada, North Dakota and Tennessee — and the District of Columbia allow self-driving cars under certain conditions.
  

Go to top

Short Features

“Our current legal system is woefully inadequate for what is coming.”

The use of robots is growing, but not the number of laws governing them. Most such laws deal with autonomous vehicles: Six states and the District of Columbia recently allowed self-driving cars under certain conditions.Footnote * California, for example, has specified that a human driver must be in the car to take over in an emergency. The Federal Aviation Administration regulates pilotless drones in domestic airspace. 1

But that's about the extent of the applicable laws, experts say.

“Our current legal system is woefully inadequate for what is coming,” says Steve Omohundro, founder of Palo Alto, Calif.-based Self-Aware Systems, which studies the safety of intelligent-machine technologies. “We haven't sorted out the basics of a legal structure for what is coming. The sooner we start on that, the better.”

Ryan Calo, an assistant professor of law at the University of Washington in Seattle, says courts have been reluctant to apply existing civil and criminal laws to machines that act while not under the control of humans. Liability for a harmful action caused by such a device can be very difficult to pin down, Calo says, because the device typically is composed of components from multiple manufacturers and users. “We have this really promiscuous data ecosystem in which information and software is flying all over the place,” he says.

“It's a very permissive, open environment, and that's great,” Calo says. “And the law has been very, very permissive about this,” particularly with respect to manufacturers' liability. “When people would lose important documents because Word froze, they wouldn't be able to go on to sue Microsoft,” he says.

With the coming generations of robots, however, Calo says the situation may change. “Where software can touch you — where it can have a physical effect,” such as a self-driving car injuring a pedestrian — “the law is not going to be able to rely upon the expedience it has had in the past,” he says.

Experts say tort laws, which establish civil liabilities for acts that harm others in some fashion, will be the first legal area that will need to accommodate robots. “If I shoot somebody, I'm clearly liable,” says Omohundro. “If I make a robot that shoots somebody, I'm probably still liable. But if I make a robot that then creates another [autonomous] robot using some complicated algorithm and that robot shoots somebody, am I liable?”

The answer is unclear, according to Calo. When robots begin hosting software and devices from various sources, he says, “the law is either going to have to decide to continue to cap liability in order to promote openness or to make the manufacturer responsible.”

The situation is especially complicated, he adds, because under tort law the potential for damage must be foreseeable. “It needs to be the kind of thing that you could have … taken steps to head off,” he says. With robots — and especially with robots that can modify their own programs — it's unclear how the manufacturer could foresee all potential cases of robot-caused damage.

“That's also going to present a major problem for criminal law, which relies upon you intending to do something in order to lock you up,” says Calo. “Criminal law … is not likely to hold you accountable for something you did not intend to do. If more and more things are happening because machines are deciding to do these things and people did not intend them …, these things are going to require rethinking the criminal law's distaste for strict liability.” Strict liability is legal responsibility for an injury, which can be imposed on the wrongdoer without proof of carelessness or fault.

Another unresolved legal issue involves semiautonomous robots: What rights, if any, should robots have when they are victims of crimes or suspects in a crime? “Right now, I don't think we have a clue as to how these new systems should be regulated,” Omohundro says. “In the longer term, I believe they are actually going to be like citizens. They will be autonomous entities with their own wants [and] desires, and they will need our legal system and our laws to apply to them.”

Omohundro adds that while some in academia are starting to grapple with the issue, “I would say legal scholars have not really begun to think about it.”

The European Union (EU) financed a two-year study of the ethical and legal issues raised by robotic applications. It concluded in 2014 with a 215-page report, “Guidelines on Regulating Robotics,” which urged EU members to harmonize their widely varying liability laws and other regulations governing robotic systems in order to eliminate legal uncertainty and thus encourage technological developments. 2

— Patrick Marshall

[1] For background, see Daniel McGlynn, “Domestic Drones,” CQ Researcher, Oct. 18, 2013, pp. 885–908.

[2] “RoboLaw,” September 2014, www.robolaw.eu/.

Go to top

Machines are handling everything from surgery to trash pickup.

The medical robot da Vinci drew considerable media attention when it began assisting surgeons after receiving approval from the Food and Drug Administration (FDA) in 2000. Since then, it has helped in a variety of soft-tissue procedures, including heart operations, hysterectomies, gall bladder and kidney removals and prostate cancer treatments. 3

Developed by Intuitive Surgical Inc., in Sunnyside, Calif., the ultrasophisticated device is among the leaders in the medical robotics market, which includes surgical robots, rehabilitation robots and robots that clean facilities and dispense medicines. A recent study valued sales of medical robots at $1.8 billion in 2013, the most recent data available, and that figure is expected to grow to $3.8 billion by 2018. 4

Hospital officials hope automation can enable them to cut costs and increase revenue. At ProMedica Toledo Hospital, in Ohio, doctors are using a machine called Sedasys that could replace highly paid anesthesiologists for colonoscopies, saving the hospital (and maybe eventually patients) money. The hospital is also considering eliminating a nursing shift in the recovery room because colonoscopy patients are recovering more quickly due to Sedasys' efficiencies, said Joseph Sferra, vice president of surgical services. Sedasys' ability to quickly sedate patients, he said, will allow the hospital to perform more procedures: The hospital could add two to three more to the current 15 per day per colonoscopy suite. 5

Critics worry about safety. The da Vinci system in 2012 helped perform more than 350,000 surgeries nationwide, a 60 percent increase from 2010. Researchers have improved the system over the years, but — just as with human surgeons — it has not performed without mishap. Indeed, the FDA received 3,697 reports of malfunctions associated with injuries or deaths in the first 11 months of 2013. 6

Doctors in Rouen, France (Getty Images/BSIP/UIG)  
Doctors in Rouen, France, use the da Vinci robot to perform a hysterectomy. Sales of medical robots were valued at $1.8 billion in 2013, a figure expected to grow to $3.8 billion by 2018. (Getty Images/BSIP/UIG)

As a result, a number of malpractice suits have been filed over surgeries involving the device. 7 Clauses to cover robots have to be included in malpractice policies.

Since 2000, a variety of robotic surgeons have joined da Vinci as automated surgery gains acceptance. Robodoc specializes in orthopedic surgery. Like da Vinci, it is controlled by a human surgeon, who programs a patient's anatomy, as well as a surgical plan, into the robot before surgery. 8

Medical automation is proceeding on other fronts that promise to transform medical care and the economics of the health industry.

Radiologists at Brigham and Women's Hospital in Boston are testing a soft robot made of plastic and ceramic to perform prostate biopsies on patients while they are inside an MRI machine. 9 The advantage of performing the biopsy inside the MRI is that real-time scans can guide the surgeon. Of course, a human surgeon can't fit inside the MRI alongside the patient, and a metal robot wouldn't work because of the MRI's strong magnetic fields.

An ingestible computer from Redwood, Calif.-based Proteus Digital Health received FDA approval in 2012. 10 The 1-square-millimeter chip is embedded in a standard pharmaceutical pill; once the pill reaches the patient's stomach and gastric juices dissolve the capsule, the chip uses those same juices as a power source to send an electrical signal that is picked up by a patch on the patient's skin. The chip is designed to transmit an ongoing real-time report of heart rate.

Researchers at Johns Hopkins University in Baltimore have created tiny robots called “microgrippers,” which are star shaped and smaller than 500 micrometers in diameter, or roughly the size of the period at the end of this sentence. The microgrippers' temperature-sensitive arms, which close when touching warm tissue, can perform biopsies. Researchers say the devices also can be used in colon cancer screenings and other procedures. 11

Scientists also are developing robotic devices to help individuals with physical disabilities. Researchers at Korea University in South Korea and Technical University of Berlin announced in August the development of a computer program that allows disabled people to control a lower-limb exoskeleton — a hardware support system worn on the lower body — with their brain waves. 12 Wearing an electroencephalogram (EEG) cap, the user can instruct the exoskeleton to move forward, turn left or right, or sit or stand by staring at one of five flickering light-emitting diodes (LEDs).

Hospitals are increasingly using robots to handle mundane tasks. At more than 400 hospitals, InTouch robots carry food trays, medical records, medications, trash and other items. Sensors and embedded GPS give the machines the ability to avoid walls, doctors, nurses and other obstacles in hallways and rooms.

The start-up firm Aethon Inc., in Pittsburgh, has sold similar robots to more than 100 hospitals. But observers say such sales are only the beginning.

“Those numbers are expected to grow quickly,” noted a Wall Street Journal reporter. “As America's elderly population grows, the country's health care system is facing cost pressures and a shortage of doctors and nurses. Many administrators are hoping to foist some of the less glamorous work onto robots. This could create a potential bonanza for software and application developers to write new programs for them, investors and industry watchers say.” 13

— Patrick Marshall

[3] Larry Greenemeier, “Robotic Surgery Opens Up,” Scientific American, Feb. 11, 2014, http://tinyurl.com/o38wy26.

[4] “Medical Robots Market by Type (Surgical Robot, Rehabilitation Robotics, Telemedicine, Assistive Robots, Orthotics, Prosthetics, Radio Surgery, Exoskeleton) & Application (Orthopedic, Neurology, Laparoscopy) — Global Forecasts to 2018,” MarketsandMarkets, March 2014, http://tinyurl.com/p22p7sg.

[5] Todd C. Frankel, “New machines could one day replace anesthesiologists,” The Washington Post, May 11, 2015, http://tinyurl.com/q3pccfh.

[6] Greenemeier, op. cit.

[7] Herb Greenberg, “Patients Scarred After Robotic Surgery, CNBC, April 19, 2013, http://tinyurl.com/c9rokd6.

[8] “Robodoc,” Curexo Technology Corp., undated, accessed Sept. 17, 2015, http://tinyurl.com/o8dwvzb.

[9] Eliza Strickland, “Inside an MRI, a Non-Metallic Robot Performs Prostate Surgery,” IEEE Spectrum, July 8, 2015, http://tinyurl.com/nu39fju.

[10] Rachel Courtland, “Medical Microbots Take a Fantastic Voyage Into Reality,” IEEE Spectrum, June 1, 2015, http://tinyurl.com/nqmmvnm.

[11] Celia Gorman, “Miniature Robots Perform Surgery,” IEEE Spectrum, June 3, 2015, http://tinyurl.com/pm7hdjn.

[12] “A brain-computer interface for controlling an exoskeleton,” IOP, Aug. 18, 2015, http://tinyurl.com/oygswjh.

[13] Timothy Hay, “The Robots are Coming to Hospitals,” The Wall Street Journal, March 15, 2012, http://tinyurl.com/og36s7l. For background, see Tom Price, “Doctor Shortage,” CQ Researcher, Aug. 28, 2015, pp. 697–720.

Go to top

Bibliography

Books

Bostrom, Nick , Superintelligence: Paths, Dangers, Strategies , Oxford University Press, 2014. A philosophy professor at the University of Oxford explores the implications of superhuman artificial intelligence and the challenges that human society will face as a result.

Brynjolfsson, Erik, and Andrew McAfee , The Second Machine Age: Work, Progress, and Prosperity in a Time of Brilliant Technologies , W.W. Norton & Co., 2014. Two MIT professors tone down their previous dire warnings about the impact of intelligent machines on the global economy and offer coping strategies for individuals and policymakers.

Carr, Nicholas , The Glass Cage: Automation and Us , W.W. Norton & Co., 2015. A New York Times reporter examines the effects of automation on society, focusing on how computers change the ways in which individuals interact with the world and with each other.

Ford, Martin , Rise of the Robots: Technology and the Threat of a Jobless Future , Basic Books, 2015. The founder of a Silicon Valley software firm cogently argues that coming improvements in robotics and machine intelligence will harm employment far more than most economists predict.

Markoff, John , Machines of Loving Grace: The Quest for Common Ground Between Humans and Robots , HarperCollins, 2015. A New York Times technology reporter offers an engaging history of the relationship between humans and machines. His analysis of the present and conjectures about the future focus on how machines can augment human capabilities.

Articles

Brynjolfsson, Erik, and Andrew McAfee , “Will Humans Go the Way of Horses?” Foreign Affairs, July/August 2015, http://tinyurl.com/p48k88x. Two MIT professors explain why in the not-too-far-off future, demand for human labor will decline; they also outline the issues society will have to grapple with to stabilize the economy.

Elliott, Stuart W. , “Anticipating a Luddite Revival,” Issues in Science and Technology, Spring 2014, http://tinyurl.com/p74wbjv. An analyst at the Organisation for Economic Co-operation and Development in Paris says the advance of robotics and machine intelligence poses a much greater technological challenge than earlier innovations.

Joy, Bill , “Why the Future Doesn't Need Us,” Wired, April 2000, http://tinyurl.com/pggy76l. The co-founder and chief scientist of Sun Microsystems delivers one of the early, and very detailed, warnings about the risks to humans of superintelligent machines.

Mokyr, Joel, Chris Vickers and Nicolas L. Ziebarth , “The History of Technological Anxiety and the Future of Economic Growth: Is This Time Different?” Journal of Economic Perspectives, Summer 2015, http://tinyurl.com/qhfrjo9. Three economists argue that while earlier technological advances didn't require government intervention, “this one will require public policy to ameliorate the harshest effects of dislocation.”

Rus, Daniella , “The Robots Are Coming: How Technological Breakthroughs Will Transform Everyday Life,” Foreign Affairs, July/August 2015, http://tinyurl.com/ndcfwkm. The director of MIT's Computer Science and Artificial Intelligence Lab says robots must be designed to fit into humans' world and not the other way around.

Reports and Studies

“RoboLaw: Regulating Emerging Robotic Technologies in Europe: Robotics Facing Law and Ethics,” FP7-SCIENCE-IN-SOCIETY-2011-1 Project No.: 289092, RoboLaw Project, September 2014, http://tinyurl.com/psw93pf. A study commissioned by the European Union examines the ethical and legal issues raised by robotic applications.

Armstrong, Stuart , “Smarter Than Us: The Rise of Machine Intelligence,” Machine Intelligence Research Institute, 2014, http://tinyurl.com/ovchl4s. This 54-page essay by a research fellow at the University of Oxford's Future of Humanity Institute explains why some experts are convinced that machines will surpass humans in intelligence and why steps must be taken to ensure human safety.

Frey, Carl Benedikt, and Michael A. Osborne , “The Future of Employment: How Susceptible Are Jobs to Computerisation?” Oxford Martin School, Sept. 17, 2013, http://tinyurl.com/oj67kae. A study by two engineering professors at a University of Oxford think tank finds that 47 percent of jobs in the United States are susceptible to being replaced by machines.

Go to top

The Next Step

Domestic Terrorism

Claburn, Thomas , “Robots, Humans Will Work Side By Side, Forrester Says,” InformationWeek, Aug. 26, 2015, http://tinyurl.com/pnyo4v5. Automation in the workplace may eliminate more jobs than it creates over the next 10 years, according to a report by the technology market research firm Forrester.

Reich, Robert , “If Technology Is Fast Making Labor Redundant, Who Can Afford to Buy Stuff?” Newsweek, Sept. 6, 2015, http://tinyurl.com/nspc2pu. Humans will have less money to spend on the goods they produce in future years as new technologies replace well-paying jobs, says a professor of public policy at the University of California, Berkeley.

Wakefield, Jane , “Intelligent Machines: The jobs robots will steal first,” BBC, Sept. 14, 2015, http://tinyurl.com/oqzvmox. Jobs in journalism, food service and medicine are at high risk of becoming automated in the next decade, according to researchers.

Laws/Regulation

Deng, Boer , “Machine ethics: The robot's dilemma,” Nature, July 1, 2015, http://tinyurl.com/pc39te5. Robotics engineers are working to develop ethical codes and enable increasingly autonomous machines to learn how to adapt to new situations.

Gray, Jeff , “If a robot kills someone, who is to blame?” The Globe and Mail, April 16, 2015, http://tinyurl.com/q4dabnn. Some legal experts question whether lethal autonomous weapons should be held liable for committing violent acts if they are designed without ethical codes.

Hiltzik, Michael , “A glimpse into the future: Will our robots have legal rights,” Los Angeles Times, Oct. 10, 2014, http://tinyurl.com/ptvzh7r. Governments in the future will likely extend legal rights and responsibilities for autonomous robots to their owners, rather than to the machines themselves, according to a professor at the University of Washington School of Law.

Robotics Industry

Alpeyev, Pavel , “Fanuc Invests in Startup as Robot Intelligence Race Heats Up,” Bloomberg Business, Aug. 21, 2015, http://tinyurl.com/pb42r48. Japanese robot manufacturer Fanuc Corp. invested $7.3 million in an artificial intelligence start-up that develops independent learning techniques for robots and enables them to share knowledge with one another via the cloud.

Markoff, John , “Korean Robot Makers Walk Off With $2 Million Prize,” The New York Times, June 6, 2015, http://tinyurl.com/nr3x4ox. Roboticists from the Korea Advanced Institute of Science and Technology won $2 million in an international robotics competition hosted by the Pentagon by designing a robot that adapts to hazardous settings.

Ramsey, Mike, and Douglas MacMillan , “Carnegie Mellon Reels After Uber Lures Away Researchers,” The Wall Street Journal, May 31, 2015, http://tinyurl.com/nefd2bg. Uber successfully recruited 40 of Carnegie Mellon University's researchers to work at the ridesharing company's nearby autonomous vehicle research facility in Pittsburgh.

Workplace Safety

Knight, Will , “How Human-Robot Teamwork Will Upend Manufacturing,” MIT Technology Review, Sept. 16, 2014, http://tinyurl.com/krw6tjl. German automaker BMW improved production efficiency and reduced worker injuries at its Spartanburg, S.C., manufacturing plant by installing cage-free robotic arms alongside human factory workers.

Masters, Brooke , “The nuts and bolts of robot-human working relations,” Financial Times, July 3, 2015, http://tinyurl.com/odvqmk9. While robotic equipment can be dangerous if mishandled, companies can improve workplace safety by installing robots that accomplish high-risk tasks and work with hazardous materials.

Go to top

Contacts

Association for Advancing Automation
900 Victors Way, Suite 140, Ann Arbor, MI 48108
734-994-6088
www.a3automate.org
Trade association for companies in the robotics industry and related businesses.

Center for Economic and Policy Research
1611 Connecticut Ave., N.W., Suite 400, Washington, DC 20009
202-293-5380
www.cepr.net
Think tank that focuses on the effects of economic policies, including automation.

Economic Policy Institute
1333 H St., N.W., Suite 300, East Tower, Washington, DC 20005-4707
202-775-8810
www.epi.org/
Advocate for low- and middle-income workers that examines the impact of robotics on the economy.

Institute for Ethics and Emerging Technologies
56 Daleville School Road, Willington, CT 06279
860-297-2376
www.ieet.org
Think tank focused on the ethics of new technologies.

International Federation of Robotics
Lyoner Str. 18, 60528 Frankfurt, Germany
+49 69-6603-1502
www.ifr.org
Trade group formed by robotics companies from 15 countries.

Machine Intelligence Research Institute
2030 Addison St., #300, Berkeley, CA 94704
https://intelligence.org/
Research organization that seeks to ensure that the creation of smarter-than-human intelligence has a positive impact.

The Robotics Institute
5000 Forbes Ave., Pittsburgh, PA 15213-3890
412-268-3818
www.ri.cmu.edu
Established at Carnegie Mellon University in 1979; conducts basic and applied research in robotics technologies.

Go to top

Footnotes

[1] Dylan Love, “Here's The Burger-Flipping Robot That Could Put Fast-Food Workers Out Of A Job,” Business Insider, Aug. 11, 2014, http://tinyurl.com/klx53px.

[2] Claire Cain Miller, “Restaurant of the Future? Service With an Impersonal Touch,” The New York Times, Sept. 8, 2015, http://tinyurl.com/oefhdhd.

[3] Patrick Thibodeau, “One in three jobs will be taken by software or robots by 2025,” Computerworld, Oct. 6, 2014, http://tinyurl.com/pypx9n7.

[4] John Tamny, “Why Robots Will Be the Biggest Job Creators in World History,” Forbes, March 1, 2015, http://tinyurl.com/psgak3g.

[5] Kirsten Korosec, “Google self-driving cars arrive in Austin,” Fortune, July 7, 2015, http://tinyurl.com/qcr93o4.

[6] Martin Ford, Rise of the Robots: Technology and the Threat of a Jobless Future (2015), p. 95.

[7] Ibid.

[8] “World Robotics 2014: Industrial Robots,” International Federation of Robotics, Sept. 30, 2014, http://tinyurl.com/owpfjv8.

[9] Ibid.

[10] Carl Benedikt Frey and Michael A. Osborne, “The Future of Employment: How Susceptible Are Jobs to Computerisation?” University of Oxford, Sept. 17, 2013, http://tinyurl.com/oj67kae.

[11] Raymond Tallis, “Conscious computers are a delusion,” The Guardian, Sept. 3, 2009, http://tinyurl.com/pvbca88.

[12] Aaron Smith and Janna Anderson, “AI, Robotics, and the Future of Jobs,” Pew Research Center, Aug. 6, 2014, http://tinyurl.com/nzjt4rz.

[13] RoboEarth, http://roboearth.org/.

[14] “Positive Impact of Industrial Robots on Employment,” International Federation of Robotics, February 2013, http://tinyurl.com/pkro74u.

[15] Ibid.

[16] Scott Andes and Mark Muro, “Don't blame the robots for lost manufacturing jobs,” Advanced Industries Series, No. 64, The Brookings Institution, April 29, 2015, http://tinyurl.com/m3s2ymx.

[17] Ibid.

[18] Ford, op. cit., p. xvi.

[19] Marc Andreessen, “This Is Probably a Good Time to Say That I Don't Believe Robots Will Eat All the Jobs,” author's blog, June 13, 2014, http://tinyurl.com/nklm9d6.

[20] Erik Brynjolfsson and Andrew McAfee, “Will Humans Go the Way of Horses? Labor in the Second Machine Age,” Foreign Affairs, July/August 2015, http://tinyurl.com/p48k88x.

[21] Ibid.

[22] Frey and Osborne, op. cit.

[23] “National Robotics Initiative,” CCC Blog, Computing Community Consortium, Nov. 24, 2014, http://tinyurl.com/q6mbnrm.

[24] “Remarks by the President at Carnegie Mellon University's National Robotics Engineering Center,” The White House, June 24, 2011, http://tinyurl.com/qxmkosh.

[25] Congressional Robotics Caucus, http://tinyurl.com/nuwzu9r.

[26] Daron Acemoglu and James A. Robinson, Why Nations Fail: The Origins of Power, Prosperity, and Poverty (2012), p. 182.

[27] Frey and Osborne, op. cit., p. 8.

[28] Joel Mokyr, Chris Vickers and Nicolas L. Ziebarth, “The History of Technological Anxiety and the Future of Economic Growth: Is This Time Different?” Journal of Economic Perspectives, Summer 2015, p. 36, http://tinyurl.com/oxojrlm.

[29] Karl Marx, Das Kapital, chapter 15, section 3; digital version available at http://tinyurl.com/qxgko7t.

[30] Nicholas Carr, The Glass Cage: Automation and Us (2015), p. 27.

[31] John Markoff, “In 1946, He Imagined an Age of Robots,” The New York Times, May 20, 2013, http://tinyurl.com/m3who6o.

[32] Ibid.

[33] John McCarthy et al., “A Proposal for the Dartmouth Summer Project on Artificial Intelligence,” Aug. 31, 1955, http://tinyurl.com/yacnlzh.

[34] Nick Bostrom, Superintelligence: Paths, Dangers, Strategies (2014), p. 7.

[35] P. W. Singer, Wired for War: The Robotics Revolution and Conflict in the 21st Century (2010), p. 58.

[36] David Conway, “Automation will create jobs in the robotics industry,” ARK Invest, Nov. 10, 2014, http://tinyurl.com/ohmm5js.

[37] “Why Robotics is the Next Big Growth Industry,” Motley Fool, April 8, 2014, http://tinyurl.com/kzdzvea.

[38] Danielle Kucera, “Amazon Acquires Kiva Systems in Second-Biggest Takeover,” Bloomberg, March 19, 2012, http://tinyurl.com/nwd726m.

[39] Steve Lohr, “IBM's Watson Attracts Commercial Clients,” The New York Times, Oct. 7, 2014, http://tinyurl.com/o7lhv5x.

[40] Lohr, op. cit.

[41] Lucas Mearian, “IBM aims Watson's cognitive computer power at CVS customers,” Computerworld, July 30, 2015, http://tinyurl.com/oajqhub.

[42] Catherine Shu, “Google Acquires Artificial Intelligence Startup DeepMind For More Than $500M,” TechCrunch, Jan. 26, 2014, http://tinyurl.com/ma6kjfg.

[43] David H. Autor and David Dorn, “How Technology Wrecks the Middle Class,” The New York Times, Aug. 24, 2013, http://tinyurl.com/oezlu3r. Also see “Will a robot take your job?” BBC News, Sept. 11, 2015, http://tinyurl.com/ntft467.

[44] Jennifer Chu, “Giving Robots a More Nimble Grasp,” MIT News Office, Aug. 3, 2015, http://tinyurl.com/phw9ylx.

[45] Larry Hardesty, “Object Recognition for Robots,” MIT News Office, July 23, 2015, http://tinyurl.com/ow2y8zk.

[46] Daniel Terdiman, “IBM's TrueNorth processor mimics the human brain,” CNet, Aug. 7, 2014, http://tinyurl.com/lmgxcnc.

[47] “DARPA SyNAPSE Program,” Artificial Brains, Jan. 11, 2013, http://tinyurl.com/d58mypl.

[48] Daniela Rus, “The Robots Are Coming: How Technological Breakthroughs Will Transform Everyday Life,” Foreign Affairs, July/August 2015, p. 2, http://tinyurl.com/ndcfwkm.

[49] Ibid., p. 4.

[50] “Global Industrial Robotics Market Forecast & Opportunities, 2018,” Research and Markets, April 2013, http://tinyurl.com/o9473m4.

[51] “Investment trend: The rise of the robots,” BNY Mellon, March 2015, http://tinyurl.com/ojjyjov.

[52] Stuart W. Elliott, “Anticipating a Luddite Revival,” Issues in Science and Technology, Spring 2014, http://tinyurl.com/p74wbjv.

[53] Gill A. Pratt, “Is a Cambrian Explosion Coming for Robotics?” Journal of Economic Perspectives, Summer 2015, http://tinyurl.com/qgbwxyr.

[54] Ford, op. cit., p. xvii.

Go to top

About the Author

Patrick Marshall, author of this week's edition of CQ Researcher  

Patrick Marshall, a freelance policy and technology writer in Seattle, is a technology columnist for The Seattle Times and Government Computer News. He has a bachelor's degree in anthropology from the University of California, Santa Cruz, and a master's degree in international studies from the Fletcher School of Law and Diplomacy at Tufts University.

Go to top



Document APA Citation
Marshall, P. (2015, September 25). Robotics and the economy. CQ Researcher, 25, 793-816. Retrieved from http://library.cqpress.com/
Document ID: cqresrre2015092500
Document URL: http://library.cqpress.com/cqresearcher/cqresrre2015092500
ISSUE TRACKER for Related Reports
Artificial Intelligence
Jul. 06, 2018  Algorithms and Artificial Intelligence
Sep. 25, 2015  Robotics and the Economy
Jan. 23, 2015  Robotic Warfare
Apr. 22, 2011  Artificial Intelligence
Nov. 14, 1997  Artificial Intelligence
Aug. 16, 1985  Artificial Intelligence
May 14, 1982  The Robot Revolution
BROWSE RELATED TOPICS:
Economic Analyses, Forecasts, and Statistics
Engineering
General Employment and Labor
International Economic Development
Manufacturing and Industrial Production
Medical Devices and Technology
Motor Traffic and Roads
READER COMMENTS
(0)
No comments on this report yet.
Comment on this Report