The AI revolution is coming, and Donald Trump isn't ready for it

The American economy is on the verge of a major transformation. Companies in Detroit and Silicon Valley are creating autonomous-driving software that promises to upend the transportation sector. These technologies threaten the jobs of hundreds of thousands of truck and taxi drivers, and they could have far-reaching implications for everything from urban planning to the retail sector.

Some economists predict this will be part of an even broader trend, as software takes over a wide range of jobs that are currently done by humans.

The Trump administration seems totally oblivious to these trends.

Two weeks ago, President Trump gave a speech at the American Center for Mobility in Detroit. I tuned in expecting he’d say something about how his administration plans to regulate the fully self-driving cars that Ford and others have vowed to introduce by 2021.

But Trump didn’t mention self-driving technology once in his speech. Instead, he promised a “new future of automotive leadership” that focused on relaxing environmental regulations and protecting American autoworkers from foreign competitors.

Then on Friday, Treasury Secretary Steve Mnuchin was asked if he was worried about artificial intelligence taking American jobs. “I think we’re so far away from that,” Mnuchin told Axios’s Mike Allen. He declared that the issue was “not even on my radar screen,” because he thought AI-driven job losses were still 50 or 100 years in the future.

Mnuchin quickly attracted criticism from experts. “If Mnuchin had done any previous reading or learning about AI, he couldn't have uttered those ridiculous words this morning,” tweeted futurist Amy Webb. She’s right.

There’s lots of evidence that AI will have a big economic impact

AI’s impact on the labor market is going to be felt a lot sooner than 50 years from now. Research suggests we’ve already been feeling it for a while. A new paper from MIT economists Daron Acemoglu and Pascual Restrepo finds that between 1990 and 2007, each industrial robot added by US industries such as auto manufacturing, electronics, and pharmaceuticals cost about 6.5 human jobs.

The overall effect was modest because the US only added about one robot for every 1,000 workers during this period. But robot sales have grown rapidly over the past decade, and some experts expect even faster growth in the coming decade.

And industrial robots represent only one area where artificial intelligence is likely to have an impact on the labor market. Self-driving cars is another. So is retail: Amazon is working to create a brick-and-mortar convenience store where cameras and AI software track which items customers take off the shelves. If widely deployed, this kind of software could cost many of the nation’s 3.4 million cashiers their jobs.

It’s too simplistic to say that these technologies will lead to mass unemployment. For one thing, new technologies often create jobs as well as destroying them. The introduction of the ATM initially increased employment of human tellers by allowing banks to open more branches.

Also, whether automation leads to lower employment or greater prosperity depends a lot on what economic policymakers — especially the Federal Reserve — do. Automation that eliminates jobs reduces manufacturing costs at the same time, and those cost savings are often passed along to consumers.

In an environment of falling prices and rising unemployment, the Fed is supposed to pump more money into the economy to boost demand and employment. So automation should only lead to mass unemployment if the Fed is asleep at the switch — or if job losses become so rapid that the central bank becomes impotent.

The Trump administration isn’t ready for the robot revolution

Photo by Jeff Swensen/Getty Images

That’s why the Trump administration’s demonstrated lack of interest in artificial intelligence is so alarming: Over the next four years, federal policymakers are going to play a crucial role in determining how AI-related breakthroughs affect the American economy and society, whether Trump realizes it or not.

A key question for Trump is who will head the National Highway Traffic Safety Agency (NHTSA), the main federal agency that regulates car safety. Trump has yet to announce his pick for the job, creating a lot of uncertainty about how self-driving cars will be regulated over the next four years. Elaine Chao, who heads the Department of Transportation, of which NHTSA is a part, has signaled that she has a broadly favorable view of the technology but hasn’t spelled out her goals in any detail.

To be fair, Barack Obama didn’t name his NHTSA administrator until April 2009, so it’s not that surprising that Trump hasn’t made a choice yet. But with Trump’s evident lack of interest in the topic, there’s a risk that the position could remain open for months to come.

The Trump administration’s attitudes stand in strong contrast to the last year of Obama’s presidency. In September 2016, NHTSA published comprehensive guidelines that laid out the administration’s vision for self-driving cars. Then in December, the administration published a report on the implications of artificial intelligence for the American economy. The report directly contradicts Mnuchin’s view that AI is decades away from having a significant impact on the US economy.

More generally, if the Trump administration is convinced that AI won’t cause job losses for 50 years or more, it is likely to misinterpret labor market trends and make policy mistakes. We can already see this in Trump’s obsession with Chinese trade — a factor that has generated real but relatively modest job losses over the past decade in the context of the overall labor market. Erecting higher trade barriers won’t do much good for American workers if technological progress, rather than trade with China or Mexico, is the primary factor destroying American jobs.

Naming a Fed chief who is committed to maintaining full employment is a key way Trump can help American workers. Expanding programs that promote entrepreneurship and job creation is another. But Trump is so obsessed with foreign trade that he may not consider other approaches.

Explanatory journalism is a public good

At Vox, we believe that everyone deserves access to information that helps them understand and shape the world they live in. That's why we keep our work free. Support our mission and help keep Vox free for all by making a financial contribution to Vox today.

One-Time Monthly Annual

$95/year

$120/year

$250/year

$350/year

Other

$

Yes, I'll give $120/year

Yes, I'll give $120/year

We accept credit card, Apple Pay, and Google Pay. You can also contribute via