Question:

1. Read the article on Roe Butt, Cy Borg, Ann Droid: Hint, They're Not Taking Your Job (written by Jason Smith) and write a one-paragraph summary of what the author's thesis is, and the research he uses to support that thesis. Pay attention to the annotations the book makes in the margins of the essay.

Article:

IN RECENT YEARS, many commentators have suggested that robots will take over most jobs in the next fifty years and leave the human workforce out in the cold. Experts even argue that this takeover is already happening. This paper challenges this alarmist way of thinking. The evidence, I argue, suggests that robots are not eliminating human jobs. In fact, they are creating new jobs. Throughout history, such innovative tools didn't end work but increased it.

To be sure, in periods when innovations take hold, workers tend to fear their own dislocation. For example, as researcher James Lacey notes, during the industrialization of the French silk industry, "French weavers threw their wooden shoes (sabots) into textile machines to make them break down. (Thus the term sabotage.)." However, it is likely that those displaced workers found other jobsas auto mechanics, gas station attendants, and car makers, for examplejust as those who worked in carriage houses and were replaced by the automobile did. Changes happen in every industry over time: the airplane replaced the train, for example, and lightbulbs replaced the need for the candlemaker, but in every era, many displaced workers found new professions. A well-defined pattern is beginning to emerge hereall technological advances bring with them new opportunities.

Predictions about the downfall of the workforce due to technological innovation have been happening for nearly a century. Jeff Borland notes that "in the 1930s, John Maynard Keynes envisaged that innovations such as electricity would produce a world where people spent most of their time on leisure." Later, during the 1930s, as Richard Freeman notes, "US President Franklin D. Roosevelt blamed unemployment on his country's failure to employ the surplus labor created by the efficiency of its industrial processes." He also points out that in the early 1960s, widespread fears that automation was eliminating thousands of jobs per week led the Kennedy and Johnson administrations to examine the link between productivity growth and employment.in the early 1960s, widespread fears that automation was eliminating thousands of jobs per week led the Kennedy and Johnson administrations to examine the link between productivity growth and employment.

Two more recent studies demonstrate contemporary arguments for the impact robots will have on our future. Race against the Machine, a book by Erik Brynjolfsson and Andrew McAfee, focuses on the results of their study of the recovery from the Great Recession of 2008. In his TED talk based on the book, McAfee uses a chart that presents a dramatic correlation between productivity and the labor force and notes that his research with coauthor Brynjolfsson reveals that from 2008 through 2011, while the US working-age population increased, the actual number of available jobs declined, even though the economy was recovering during that time frame and productivity actually increased (00:0:51-1:56). Their inevitable conclusion, according to McAfee, is that robots are already replacing the human workforce.

However, the data gathered in Race against the Machine is shortsighted. Had the study continued only three more years, the authors would have seen the graph's job curve return to normal values. After any recession, the rebound in employment has always taken longer than the rebound in the GDP, for several reasons. Manufacturers reduce their inventories, which makes it look like fewer people are producing more. In a recession, the remaining workers always find ways to pick up the slack left behind by laid-off workers, often through the use of overtime, which is cheaper in the short run than hiring new workers or rehiring those laid off. Once again, the greater productivity stems from existing workers, not automation. The most recent recession, moreover, produced some unusual employment data. Part-time workers are not counted as full-time employees, yet their work output is included in the overall productivity numbers.

A second alarming view of robots taking home the paycheck comes from Michael Chui and colleagues in a McKinsey Quarterly article titled "Where Machines Could Replace Humansand Where They Can't." They write:

Last year, we showed that currently demonstrated technologies could automate 45 percent of the activities people are paid to perform and that about 60 percent of all occupations could see 30 percent or more of their constituent activities automated, again with technologies available today.

The problem with this report is its use of probability for its logic: just because a thing can be done does not mean it will be done. As Gillian White points out in her reporting on a study by two economists, "The authors note that just because an industry can automate doesn't mean that it will." Costs in every case must be examined, as well as return on investment. Theoretical predictions only go so far; there is a real-world cost/benefit analysis that must be applied to each of these possible robot advancements. In addition, Chui and colleagues overlook arguments that it will require more humans to maintain and build robots than they replace (Metz).

In fact, convincing, well-researched papers that examine data over a much greater time span find there is no evidence of a robot takeover. These studies indicate that the opposite is true. For example, Jeff Borland and Michael Coelli, two Australian researchers, published an exhaustive study in 2017 that found that

(i) the total amount of work available [in Australia] has not decreased following the introduction of computer-based technologies; and (ii) the pace of structural change and job turnover in the labor market has not accelerated with the increasing application of computer-based technologies. A review of recent studies that claim computer-based technologies may be about to cause widespread job destruction establishes several major flaws with these predictions. (377)

Borland and Coelli found that fears about job losses have happened often throughout history: "Our suggested explanation for why techno-phobia has such a grip on popular imagination is a human bias to believe that 'we live in special times'" (377). In other words, we think that our particular era is uniquethat now is truly the time that robots will take over our jobsbut such a scenario doesn't come to pass.

There are more practical reasons why robots will not replace humans in many fields, especially those areas of work that require direct interaction with humans. In his Tech.Co article "5 Reasons Why Robots Will Never Fully Replace Humans," Lanre Onibalusi notes that robots cannot "understand irrational thought," cannot "understand context," and "lack creative problem solving" abilities and that "people prefer to talk with a human." Additionally, as Ryan Nakashima reports, behind almost every form of robotization is an army of technocrats. This army of workers is considered the "dirty little secret" behind the artificially intelligent robot. Whether it be the hundreds of programmers or the vast numbers of call-center operators required to carry the load, robots require human helpers. It seems that robots simply cannot understand context, and enormous numbers of analysts are required to try to reteach an automated system every time it fails to understand a request from a customer (Nakashima). In fact, according to a recent New York Times article, grocery retailers who have started using robots report that "the robots are good for their workers. They free up employees from mundane and sometimes injury-prone jobs like unloading delivery trucks to focus on more fulfilling tasks like helping customers" (Corkery). In this case, as in others, robots help increase productivity rather than replace humans.

Some of these alarmists argue that as robots become smarter, laws must be instituted to guard against a massive upheaval, or even elimination, of the human workforce. The problem with this preemptive approach is that no one can possibly imagine where or when those laws should exist or how to apply them. The normal course of events is to first have solid evidence that some event is "bad" and later proceed to legislate. For example, those who oppose self-driving cars, another kind of robot, call for governmental restrictions on them, but before such vehicles are on the road, there's no reason to fear them. In fact, they require human help. Many have heard of the self-driving truck that hauled two thousand cases of beer 120 miles in Colorado without a driver at the wheel (Isaac). What was not revealed was that there was a full police escort for the entire route and that the vehicle was incapable of negotiating the on- and off-ramps. I will admit that during the writing of this paper, a self-driving-car fatality did occur when an Arizona woman was struck and killed by a driverless vehicle owned by Uber. Uber immediately suspended all driverless test vehicles in the United States and Canada, and the following week Nvidia suspended all tests with the same product (Sage). Perhaps here is an example of a worthy event triggering the possibility of new laws, but an overall ban on automation itself in the workforce seems hasty.

As Jeff Borland has astutely observed, "The tale of new technologies causing the death of work is the prophecy that keeps on giving." It is sensational, alarmist, and grossly premature. I agree with those who believe society should not worry about robots taking away all the jobs. Most of the predictions for the future are countered by past facts. The reality is that all new technologies bring about new avenues of invention, experimentation, and work and that much of the new technology requires the intervention of humans. It is therefore unnecessary for governments to regulate robots to protect jobs.