This past October, the Pew Research Center released Automation in Everyday Life, a report on what Americans think about advanced technologies like artificial intelligence and robotics, and the impact they expect them to have on their everyday lives. The report is based on a national survey of 4,135 randomly selected American adults. To help gauge their opinion on such complex topics, the survey questions were framed around four specific scenarios related to these advanced technologies: driverless cars, workplace automation, robot caregivers, and computer algorithms that evaluate and hire job applicants.
“Americans anticipate significant impacts from various automation technologies in the course of their lifetime,” noted the report. “Although they expect certain positive outcomes from these developments, their attitudes more frequently reflect worry and concern over the implications of these technologies for society as a whole.”
Let me discuss some of the report’s key findings.
Americans were particularly concerned about a future where robots and computers can do many human jobs, with 72% expressing worry while 33% were enthusiastic. If forced to compete with machines for jobs, 76% fear that economic inequality will become much worse, 75% believe that the economy will not create enough better paying jobs, and 64% expect that people will have a hard time finding meaningful things to do with their lives.
They were also three times as likely to be worried (67%) than enthusiastic (22%) if hiring decisions are made by AI algorithms without any human involvement. On the other hand, respondents held more balanced views toward robot caregivers, - 47% worried versus 44% enthusiastic; and toward driverless vehicles, - 54% worried versus 40% enthusiastic.
It’s not at all surprising that the top concern expressed is the impact of automation on jobs. Automation anxieties have continuosly surfaced over the past two centuries whenever a major new technology came along. Given our increasingly smart machines, such concerns have understandbly accelerated in recent years. When machines do everything, what am I going to do?; Will a robot take my job away?; How are humans going to make a living?; Will my children be better off than I am?
As a 2014 Economist article reminded us, “Previous technological innovation has always delivered more long-run employment, not less… Yet some now fear that a new era of automation enabled by ever more powerful and capable computers could work out differently.” Will this time be different, as AI-based innovations end up replacing a large portion of the workforce? We don’t really know.
The majority of Americans are reluctant to incorporate these types of technologies into their own lives, concerned about removing the human element altogether
Despite acknowledging their potential benefits, almost 59% of respondents would not want a robot caregiver to take care of them or a family member. 70% agree that robot caregivers would help alleviate the burden of caring for aging relatives, but 64% worry that these technologies would increase feelings of isolation for older adults.
Similarly, 56% would be reluctant to ride in a driverless vehicle. 75% agree that driverless vehicles would help the elderly and disabled live more independent lives, while 39% expect that they would reduce the number of people killed or injured in traffic accidents. Yet, 30% said that driverless vehicles would make the roads less safe for humans.
“Those who are hesitant to use these technologies frequently describe their concerns as stemming from a lack of trust in technological decision-making and an appreciation for the unique capabilities and expertise of humans.” This lack of trust is quite understandable.
As MIT professor Sandy Pentland recently wrote, “there is a deep mistrust of user-facing automation and automatic AI systems. As a consequence, capabilities that can reduce the human-intensive nature of operations go unused… AI systems are often not used because they are built as black boxes and do not provide transparency into what assumptions and decisions the underlying algorithms are making on the user's behalf. For humans to effectively team with automation, displays are needed that can explain and visualize what decision criteria the system uses. Recent studies suggest that this transparency can improve user trust and automation adoption.”
There’s broad public support for policies that limit the scope of automation technologies and bring humans more fully into the process
The vast majority of respondents would support policies that limit these advanced technologies to specific situation, while bringing humans more fully into their operations. For example, 85% of favor limiting fully autonomous machines to those jobs that are dangerous or unhealthy for humans.
In addition, if computers and robots were capable of doing many more human jobs, 60% would be in favor of a guaranteed income, while 58% would support a national service program that would employ people to perform specific jobs. 58% of respondents also believe that there should be some limits on the number of jobs companies can replace with machines.
Americans are strongly in favor of having humans more involved with these technologies. 87% would favor requiring that all driverless vehicles have a human in the driver’s seat ready to take control as needed. 48% would feel better about a robot caregiver if a human operator could remotely monitor its actions. And, while 76% would not apply for a job that used a computer program to evaluate and select applicants, 57% would support the concept if the algorithms are only used for the initial screening of candidates, with the final selection made in a traditional in-person interview.
Many Americans expect a number of professions to be dominated by machines within their lifetimes - but relatively few expect their own jobs to be impacted
Most Americans believe that the very nature of work and jobs will be significantly transformed over the next several decades. 77% say that robots and computers will one day automate many of the jobs currently done by humans. However, only 30% expect that their own jobs will be automated during their lifetimes.
Other recent studies reached similar conclusions. A recent survey conducted by Burson-Marsteller found that while only 33% said that the country is headed in the right direction, 67% said that they were somewhat or very optimistic about the future. To a large extent, this optimism was based on the belief that their jobs are safe over the next 5 years, with 62% believing that they won’t be laid off and 60% that a machine couldn’t replace them.
A 2016 Pew Research study, The State of American Jobs, found that most Americans were happy and secure in their jobs. Almost 80% said that they were satisfied with their current jobs, and almost 90% said that they were not likely to lose their jobs in the next 12 months. But, less educated workers felt more vulnerable. They were more likely to believe that their current skills are insufficient to advance in their careers, while also feeling less secure about their current jobs. Almost 40% of those without a high school education said it was very or fairly likely that they might be laid off within the next 12 months, while only 7% with a bachelor’s or higher degree said the same thing.
Workers lacking a college education are much less likely to express positive attitudes towards the current generation of workforce technologies
These recent surveys also agreed that the current generation of workforce technologies has had widely disparate impacts on workers. “For some – especially those with high levels of educational attainment – technology represents a largely positive force that makes their work more interesting and provides opportunities for career advancement,” notes the 2017 Pew Research survey. “But those who have not attended college are much less likely to view today’s workforce technologies in such a positive light.”
53% of college graduates and 51% of those with some college felt that technology increased their opportunities for advancement, while only 32% of those with high school diplomas or less felt the same way. And, while 64% of college graduates and 54% of those with some college felt that technology makes their jobs more interesting, only 38% of those with high school or less agreed.
Similarly, the Burson-Marsteller survey found that feelings about the future differed by educational level. 71% of Americans with a college education or more said that they have the right skills to succeed in the 21st century, compared to 42% of those with a high school education or less. Only 14% of those with a college education worried that their jobs could be automated within five years, compared with 30% of those with high school or less.
The public anticipates widespread advances in the development and adoption of automation technologies over the coming decades
Finally, Americans anticipate that these various advanced technologies will make significant inroads in the coming decades. “Driverless vehicles are perhaps the most prominent example of this trend… two-thirds of the public anticipates that most vehicles on the road will be driverless within the next half-century – with 9% predicting that this will occur in the next 10 years… Additionally, substantial shares of Americans think it’s likely that within 20 years doctors will rely on computer programs to diagnose and treat most diseases (79% think this will definitely or probably happen); that most retail interactions will be fully automated and involve little or no human interaction between customers and employees (65%); and that most deliveries in cities will be made by robots or drones instead of humans (65%).”