Robots will not take as many jobs away from humans as feared, but we're still likely to face serious social and political challenges driven by the economic effects of technological change.
In recent years, a number of academic researchers have raised the possibility that advances in artificial intelligence and related technology will allow many jobs to be automated, leading to widespread unemployment and social unrest. Perhaps the most widely noted report on the subject, "The Future of Employment: How Susceptible Are Jobs to Computerisation?," came from Oxford University's Carl Benedikt Frey and Michael Osborne in 2013. MIT's Erik Brynjolfsson and Andrew McAfee have also explored the topic in The Second Machine Age, among other works.
However, a new Organisation for Economic Co-operation and Development (OECD) report, "The Risk of Automation for Jobs in OECD Countries: A Comparative Analysis," by Melanie Arntz, Terry Gregory, and Ulrich Zierahn, finds that there will be jobs for people in the future, but still foresees difficulties, particularly for low-skilled workers.
"The main conclusion from our paper is that automation and digitalization are unlikely to destroy large numbers of jobs," the authors state in the report, released this month. "However, low qualified workers are likely to bear the brunt of the adjustment costs as the automatibility of their jobs is higher compared to highly qualified workers. Therefore, the likely challenge for the future lies in coping with rising inequality and ensuring sufficient (re-)training, especially for low qualified workers."
Frey and Osborne estimated that 47% of jobs in the US could potentially be automated. If that happened over a short period, unemployment would almost certainly lead to widespread social unrest and political turmoil. But Arntz, Gregory, and Zierahn estimate that only about 9% of jobs on average across OECD states are automatable.
Their estimate varies by country. In South Korea, for instance, it's 6% while in Austria it's 12%, based on differences in workplace organization, automation investments, and worker education levels. Variations aside, the authors argue, the assumption that whole jobs will be automated rather than specific job-tasks leads to an overestimation of job losses, "as occupations labelled as high-risk occupations often still contain a substantial share of tasks that are hard to automate."
In the past, computerization has tended to lead to a change in job tasks rather than a change in employment share between occupations, the authors claim. They also observe that just because something can be automated doesn't mean it necessarily will be automated.
That conclusion can be seen in the continued presence of Starbucks and other coffee houses, or in commercial air travel. We have the technology to dispense coffee from a machine or to fly planes by autopilot. But we still prefer to buy coffee from people, and we find comfort in human pilots, despite occasional crashes attributable to pilot error or pilot malfeasance.
Legal and ethical obstacles will also slow the advance of automation, the authors say. Thus, self-driving cars are likely to be ready on a technical level before society is prepared to accommodate them.
While workers as a whole will not be made obsolete by automation, the authors assert that workers without much education will bear the brunt of the workplace change, which will necessitate investments in occupational training programs to help people adapt.
"This study clearly points towards the need to focus more on the potential inequalities and requirements for (re-)training arising from technological change rather than the general threat of unemployment that technological progress might or might not cause," the authors conclude.
(Cover image: 3alexd/iStockphoto)