Study: Workers in heartland states most at risk of losing jobs to AI

People who live in heartland states such as Kentucky and Indiana are most at risk in the United States of losing their jobs to robots or other artificial intelligence, according to a new report published Thursday from the Brookings Institution.

President Donald Trump carried many of these states in the 2016 election with promises to improve job opportunities for these "forgotten" Americans he said were left behind as other parts of the country recovered from the 2008 financial crisis.

But the think tank found an impending wave of artificial intelligence could disrupt jobs even more in these states. Repetitive tasks that involve processing information, performing physical activities or operating machinery will be the first to be replaced by artificial intelligence - which could hit manufacturing jobs hard.

"The next 30 years and the transition from the computer era into the AI era will continue to affect manufacturing," said the Brookings report's co-author Rob Maxim. "It still hasn't hit the bottom."

The problem is so massive, researchers say the federal government needs to address it - and fast. "It's urgent. I think it's something Congress should have worked on yesterday," Maxim tells me.

While the idea of tech replacing workers jobs' may seem abstract, the Brookings findings take on increasing importance as experts in Washington and on Wall Street alike are worried about the prospect of a looming recession. Since companies tend to invest more in technologies that cut labor during economic slowdowns or downturns, even more jobs could be lost in the coming years.

This leaves a prime opening for Congress to address the problem - particularly through bills that would provide more funding for programs that retrain displaced workers and offer financial assistance between jobs. Workers are more likely to participate in new trainings or certifications during a downturn, too. "It will need a very productive federal government effort as well as state and local efforts," Maxim says.

However, that's easier said than done. It seems unlikely Congress will be especially proactive on a problem that would require a major societal shift - given its inability to solve the most urgent problems right in front of it, such as reopening the government.

But the report, which emphasizes the states and demographic groups that would be most impacted by the rise of AI, could give politicians some incentives to act. After all, several states in the Midwest are expected to be a key battleground in the 2020 election. While Democrats have primarily led the charge in creating policies that would support workers displaced by automation, Republicans have skin in the game on this issue also because geographically, many of the workers most vulnerable to AI are in the rural and Midwestern states that went for Trump in 2016.

The researchers calculated a score that measured each state's exposure to automation based on existing research about what jobs are most likely to be replaced by machines. Heartland states were most at-risk, while states like Massachusetts and New York were the safest. The gap was even wider when the researchers drilled down further into states and found rural communities stood to see more jobs displaced by AI than urban areas.

They also found Hispanic and American Indian workers are more likely to be replaced by machines due to occupational segregation. Hispanic and American Indian workers, who are more likely to work in fields like construction, were far more at-risk than white and Asian workers. The research underscores that this is a trend that should concern both political parties because it hits demographics that are key to both.

Despite the research, many in the tech industry are skeptical that lawmakers will take a proactive approach when it comes to artificial intelligence. Many technologies such as food prep and delivery robots are already here, and other technologies such as self-driving trucks aren't likely too far off.

The Brookings researchers say there's no "silver bullet policy" that can address the threat artificial intelligence poses to many positions, but they recommend that policymakers consider initiatives that would invest in re-skilling workers and expand accelerated learning and certifications.

They also say policymakers should create a "Universal Adjustment Benefit" that would support workers that are displaced with retraining and financial support during extended unemployment. Already, there are limited programs like this available for workers, especially who are displaced because of trade policies. But the study's authors believe such programs need to be expanded quickly to cover workers who are displaced by new technologies.

Efforts to expand such protections have largely faltered in Congress. Last summer, Sen. Kirsten Gillibrand, D-N.Y., introduced legislation that would support Americans displaced by automation with job retraining and financial assistance. The TAA for Automation Act of 2018 stalled in the Senate. As Gillibrand enters the 2020 race, she could potentially elevate the issue. She mentioned job training programs as a key issue when she announced she was forming a committee to explore a presidential campaign on "The Late Show With Stephen Colbert."

Another effort in Congress has been the Artificial Intelligence Caucus. Members of the caucus have introduced legislation that would create a formal federal panel on artificial intelligence that would study the impact it will have on jobs. Rep. Jerry McNerney, D-Calif., was recently named the caucus co-chair in December.

"The rise of artificial intelligence offers a host of new benefits for Americans - including new job opportunities," McNerney told me in a statement. "However, as our economy shifts, it's critical that we examine the impact of AI on our nation's workforce and potential displacement that may result."