Recent reports of strong job growth in the US suggest the economy may currently be at “peak employment.” That fact has made the widely-publicized concerns over the impact of automation and artificial intelligence on jobs seem overblown and out of step with the times. Yet, the clouds of the next economic contraction are gathering.
Some economists and executives believe a recession could come as early as this year while others expect it’s another year or two out. Rana Foroohar, a global business columnist and an associate editor at the Financial Times, recently stated on CNN that more than half of CFOs at large U.S. companies expect a recession in the next year.
Might the AI worries about jobs come to fruition during this next recession?
Even though the statistics say that the economy is at full employment, this is misleading. Unemployment may be low, but underemployment is far higher. Though the U.S. Bureau of Labor Statistics does not provide an official underemployment number, a Newsweek story claims fully 20% of men aged 24 to 55 do not have full-time jobs. An NPR story further reports that around 33% of college graduates are underemployed. And this is during what is widely perceived as a boom time.
Here’s the point: Even in a strong economy there is underlying job weakness. Automation, the “gig economy” and AI are likely all contributors to underemployment and the sense of unease. The next recession will accelerate the impacts of these trends.
Nobel laureate economist Wassily Leontief issued this warning in 1983 on the introduction of increasingly sophisticated computers: “The role of humans as the most important factor of production is bound to diminish in the same way that the role of horses in agricultural production was first diminished and then eliminated by the introduction of tractors.”
That viewpoint has gained steam over the last few years regarding the job-destroying potential of AI. A Pew Research study of “experts” found that 48% “envision a future in which robots and digital agents have displaced significant numbers of both blue- and white-collar workers, with many expressing concern that this will lead to vast increases in income inequality, masses of people who are effectively unemployable, and breakdowns in the social order.”
McKinsey estimates that between three and 14% of the global workforce will need to switch occupations over the next 10 to 15 years. There is also potential for unpredictable consequences from these shifts. A Brookings Institute report cites Pew and several other studies to conclude 38% of current jobs could be automated. Due to the potential for significant job losses, their perspective is that “Western democracies likely could resort to authoritarianism as happened in some countries during the Great Depression of the 1930s in order to keep their restive populations in check.” Brookings is hardly alone in this dim view. In a 60 Minutes interview Kai-Fu Lee, one of the world's foremost experts on artificial intelligence, claimed that in as soon as 15 years the technology could displace about 40% of the jobs in the world..
What’s more, the time between now and 2025 is only the first phase of what will be a dramatic shift. A further study from PwC describes three overlapping cycles of automation that will stretch into the 2030s, each with their own degree of job impact. In a further report from McKinsey, the firm notes that automation and AI will lift productivity and economic growth, but millions of people worldwide may need to switch occupations or upgrade skills.
There are others who take a sanguine view and believe AI is more likely to increase the number of jobs. The optimistic argument is that technology revolutions always result in job displacement but ultimately create new industries and opportunities, even if it is difficult from the current perspective to envision what those industries might be, and those industries will have to hire people. That has been the story with technology revolutions so far, and is well articulated in an article appearing in Singularity Hub. In this, the author claims that “AI will be the greatest job engine the world has ever seen.”
This view is mostly consistent with a World Economic Forum report that claims a net positive outlook for jobs between now and 2022, noting that “75 million jobs may be displaced by a shift in the division of labor between humans and machines, while 133 million new roles may emerge that are more adapted to the new division of labor between humans, machines and algorithms.” Similarly, a PwC study asserts that AI, robotics, and related technologies should “generate enough new jobs to broadly offset the potential job losses associated with automation.” This view is echoed by analyst firm Gartner, which stated that AI-related job creation will reach a net 2 million new jobs by 2025.
Next recession to usher in a wave of AI
Optimistic perspectives on AI job impact may ultimately prove true in a macro sense. But along the way, many will be displaced by AI-driven automation. After all, as noted in the Newsweek story, AI and automation will replace most human workers because they don’t have to be perfect, just better than you. A major difference with the AI revolution compared with earlier technology shifts is that the impact will come much faster.
An article in The Parliament states the next recession will bring with it a wave of AI. In this, London School of Economics Professor Mirko Draca said this new wave of automation was likely to take root during the next major economic downturn, when technology-inspired restructuring in jobs tends to take place. In a report that he co-authored, Draca noted as an example the impact of autonomous vehicles that he claims could lead to the annual shedding of around 200,000 drivers from the 3.5 million-strong US commercial driver workforce.
An uncertain future
While some displaced workers may ultimately gain retraining and new opportunities, others will be left behind. The numbers of the displaced could be significant and a reactionary backlash can be expected. Foreshadowing this possibility, there are already incidents of people attacking self-driving vehicles including at least 21 leveled at Waymo self-driving vans in Chandler, Arizona as first reported by The Arizona Republic. So far, the perpetrators have not been captured but as these and future responses escalate it is easy to imagine deployment of a sophisticated infrastructure of surveillance and AI similar to what is being used today in Xinjiang province in China ostensibly to combat separatists.
The impact of the AI-powered employment transition will almost certainly have far-reaching social and political implications. Futurist Martin Ford speculated in a TedTalk about AI that we could very well end up with significant unemployment and “soaring levels of inequality.” To help offset this, he and others believe it is time to seriously consider providing people with a guaranteed income or universal basic income. Perhaps too, automation and AI is a contributing a further polarization of political views and is possibly a contributing reason behind the recent increase in the popularity of Socialism, especially among young Americans who value universal health care, free public college and living wages.
Going forward, there is really no certainty about the jobs outlook in an era where machines will increasingly do the work for us. What is clear is that the transition to the AI economy will be uneven and many people will be left behind. The next recession will provide a taste of that future.
Gary Grossman is Senior Vice President and Technology Practice Lead, Edelman AI Center of Expertise.