How quickly will AI eat our jobs? And can people stay ahead of that?
New Pew report focuses on reskilling people displaced by AI and robots
The impact of AI on the future of work and jobs can be condensed into two far-ranging questions: Will well-prepared workers be able to keep up in the race with AI tools? And will market capitalism survive?
On 3 May, the Pew Research Center released a widely reviewed report, The Future of Jobs and Jobs Training, researched and written by Lee Rainie and Janna Anderson, which focuses on exploring those questions. It may be too soon to conclusively answer them.
I was prepared for the report’s release, since I participated as one of the contributors to the research, and have the honor to serve on Pew’s advisory board for AI, Robotics, and the Future of Work.
Aside from the report’s broad survey and the contrasting viewpoints of the many contributors, the preamble brings into high relief the stakes of the automation and ephemeralization of work. The authors enumerate the myriad occupations being threatened, and then they summarize in this way:
Multiple studies have documented that massive numbers of jobs are at risk as programmed devices — many of them smart, autonomous systems — continue their march into workplaces.
Expert opinion on the threat of AI to jobs is split. This is not climate change, where the overwhelming majority of experts are united in their views.
When Pew Research Center and Elon University’s Imagining the Internet Center asked experts in 2014 whether AI and robotics would create more jobs than they would destroy, the verdict was evenly split: 48% of the respondents envisioned a future where more jobs are lost than created, while 52% said more jobs would be created than lost. Since that expert canvassing, the future of jobs has been at the top of the agenda at many major conferences globally.
I was a contributor in that 2014 Future of Jobs study and said this at the time:
I stand by that prediction, although I now think it will be the central question of 2020, which for all intents and purposes is the day after tomorrow, at least in terms of policy formulation at the national and international levels.
In the new report, I am among the 30% of respondents that don’t believe present or near-term training systems will be adequate to retrain the majority of the workers that will be displaced in the next decade.
On balance people who went back to school at the small technical college were less likely to have steady work all seasons of the year compared to people who had not retrained. And if you looked at their income from before the Great Recession to then, their income drops were greater. — Amy Goldstein
That sobering view is supported by a great deal of evidence, like that recounted by Pulitzer-prize winning journalist Amy Goldstein, whose book Janesville recounts the cascade of bad outcomes for Janesville Wisconsin after the GM factory was shut in 2008. She was interviewed by Ray Suarez of PBS in April:
SUAREZ: One of the most difficult parts to read was how upskilling and retraining and going back to community college didn’t really change the outcomes for a lot of these workers.
GOLDSTEIN: It’s a very popular idea that people who lose work should go to school and reinvent themselves. And it’s striking that it’s one of the few economic ideas in which both Democrats and Republicans tend to put stock. So I wanted to look at what was really happening. If you think about it, factory workers going back to school having lost their jobs, I mean — they were scared about studying. They were scared about how they were going to put dinner on the table. They were scared about what was going to come next. I mean, this was a pretty traumatic time.
And for these people to have to start studying on top of it all, it was a pretty big challenge. So what I looked at with the help of a couple of good labor economists was what did the data show? And it turned out that, you know, not today, but if you looked a few years out from when these jobs went away, on balance people who went back to school at the small technical college were less likely to have steady work all seasons of the year compared to people who had not retrained. And if you looked at their income from before the Great Recession to then, their income drops were greater.
Of course, these are Wisconsin factory workers, not college-educated metropolitans, and we can expect a wide range of responses to retraining. However, retraining is clearly not going to be a panacea. At least that is my conviction.
The most common response to the question from the 70% on the other side is that humans have always responded to technology disrupting traditional work by creating new sorts of jobs, like the shift to industrial production in the U.S., when former agricultural workers migrated to manufacturing centers and went to work in the factories. For example, Jonathan Grudin, principal researcher at Microsoft, commented,
People will create the jobs of the future, not simply train for them, and technology is already central. It will undoubtedly play a greater role in the years ahead.
The themes that emerged in the research lean toward a series of policy decisions that our political leaders should be grappling with, but are not key parts of today’s most heated political discourse. Pew isolated these 5 themes, the first three hopeful, and the last two less so:
- The training ecosystem will evolve, with a mix of innovation in all education formats
- Learners must cultivate 21st‑century skills, capabilities and attributes
- New credentialing systems will arise as self-directed learning expands
- Training and learning systems will not meet 21st-century needs by 2026
- Jobs? What jobs? Technological forces will fundamentally change work and the economic landscape
On one hand, optimists believe we can scale up to retrain millions of workers, like Justin Reich of Rensselaer Polytechnic Institute:
Educators have always found new ways of training the next generation of students for the jobs of the future, and this generation will be no different.
Or Ben Schneiderman of the University of Maryland, who like others seemed to be projecting a business-as-usual, traditional college education mindset when he wrote this:
Students can be trained to be more innovative, creative and active initiators of novel ideas. Skills of writing, speaking and making videos are important, but fundamental skills of critical thinking, community building, teamwork, deliberation/dialogue and conflict resolution will be powerful. A mindset of persistence and the necessary passion to succeed are also critical.
I remain unconvinced because I am trying to imagine millions of unemployed workers — former x-ray technicians, taxi drivers, financial reporters, lawyers, fast food staff, stock traders, and insurance claims adjusters — of all ages, in all geographies, seeking to get regrooved and back to work. And many of them might respond like the GM workers in Janesville.
On the more pessimistic side, without even mentioning the scale of effort involved, I stated in the new Pew report:
While we may see the creation and rollout of new training programs, it’s unclear whether they will be able to retrain those displaced from traditional sorts of work to fit into the workforce of the near future. Many of the ‘skills’ that will be needed are more like personality characteristics, like curiosity, or social skills that require enculturation to take hold. Individual training — like programming or learning how to cook — may not be what will be needed.
I’ll leave this with just one additional data point, which is recent news. In March 2017, Laurence Fink, the chief of BlackRock, the largest investment house in the world, announced they would be consolidating a number of funds — with $30 billion in assets — and laying off the majority of the quants managing those investments, shifting control to AI:
“The democratization of information has made it much harder for active management,” Mr. Fink said in an interview. “We have to change the ecosystem — that means relying more on big data, artificial intelligence, factors and models within quant and traditional investment strategies.”
As part of the restructuring, seven of BlackRock’s 53 stock pickers are expected to step down from their funds. Several of the money managers will stay on as advisers. At least 36 employees connected to the funds are leaving the firm.
Later that same week, the US Treasury Secretary, Steve Mnuchin, told an audience in Washington,
In terms of artificial intelligence taking over the jobs, I think we’re so far away from that that it’s not even on my radar screen. I think it’s 50 or 100 more years.
That disparity in outlook — where a CEO is taking active steps to embrace AI to improve the company’s bottomline by laying off dozens of highly-paid and highly-trained financial analysts, while one of the most important figures in government states that AI is decades away from posing any serious threat to jobs — should be a greater concern than our projected ability to scale up to retraining millions of displaced workers.
Without a widespread awareness of the transformation already taking place at the foundation of our economy, the perspectives of the business sector and the public sector are diverging, and the average worker might fall in the trench between them. The entire country may become Janesville, and it could be coming sooner than Mnuchin and the techno-utopians believe.
(Want to get insights into emerging tech on a more regular basis? Sign up for the official Traction Report newsletter here.
Interested in AI? Request an invite for the upcoming Velocity Network breakfast on AI and Machine Learning, May 25th in NYC.)
