SKILLS BLOG

AI in the labor market: A Jump Ball Moment for the Middle Class

By Brooke DeRenzis, Chike Aguh, August 28, 2024

If you follow basketball, you know the feeling of anticipation as the referee throws the ball up and two opposing players jump for it. As the ball spins in the air, it is anyone’s game. But once someone gains possession of the ball, their team gets to set the pace.

We are at a jump ball moment when it comes to AI’s impact on the labor market. We can either use AI to foster economic opportunity for people without a bachelor’s degree and grow the middle class, or we can let it widen existing inequalities and leave workers sitting on the bench.

The trajectory is up to us as a nation. Most experts predict that AI will create new jobs, augment existing ones, and eliminate others. But how and to what extent depends on decisions that leaders make now. With the right set of public policies, workforce policies included, our nation can ensure humans and AI are playing fair together and creating outsized impact for working people and businesses.

The American public is looking for elected officials to get in the game and take action. Two out of every three voters believe that AI is a serious or very serious problem facing American workers, and nearly the same number (64%) are concerned that workers have limited opportunities to access skills training that could help them adapt to AI in the workplace. These concerns cut across race, political party, age, and job type, but are particularly pronounced among voters without a bachelor’s degree.

So what can policymakers do now to address these concerns and ensure that working people and the businesses that employ them can thrive in the age of AI?

To start, policymakers can make sure that everyone has access to digital skills, including skills required to use AI. Research by National Skills Coalition and the Federal Reserve Bank of Atlanta finds that while 92 percent of jobs require digital skills, one-third of workers don’t have the foundational digital skills necessary to enter and thrive in today’s jobs. The digital skill divide disproportionately impacts people of color, rural residents, and people with low incomes due to historic underinvestment and structural inequities. Without expanded access to digital and AI skills, AI will only exacerbate the digital divide.

Policymakers can immediately expand people’s access to AI skills by investing in high-quality training programs so that workers can build and upgrade the digital skills necessary to work alongside AI and digital technology. The bipartisan Digital Skills for Today’s Workforce Act introduced in Congress earlier this year is one such example of proposals being put forward.

Policymakers can also take action to ensure that people most familiar with the work – workers themselves – are able to provide input on how AI is developed and deployed. Workers have valuable knowledge about the nature of their jobs, and their expertise can inform how technology is used so that it complements human talent, increases productivity, and mitigates harm. The more inclusive the development of AI is, the less likely it is to reproduce biases.

Policymakers can help people have a say in how AI is used in the workplace by protecting workers’ rights to organize. Research shows that job standards and safeguards championed by labor yield advantages for all working people, not just those in unions. In the age of AI, honoring workers’ rights and voices can also help companies safely and equitably unlock innovation and maximize the benefits of new technology. Just look at the first of its kind tech-labor partnership on AI between Microsoft and the AFL-CIO.

Workers’ perspectives should also shape AI training and upskilling programs so they are responsive to their needs. To support this, policymakers can invest in industry partnerships that bring together local businesses, unions and workers, education and training providers, and community-based organizations to develop comprehensive workforce strategies for a local industry. Through these partnerships, workers can give input on training and curriculum so that it is most relevant to their day-to-day work. The California High Road Training Partnership program and Real Jobs Rhode Island are two examples of how states have invested in industry partnerships that benefit workers and businesses. Other states, along with the federal government, can do the same.

Finally, we need better policies to support workers who are displaced from their jobs due to AI, or other structural changes to the economy. In 2020, the COVID-19 pandemic shined a light on the need for the U.S. to have a better system for supporting workers who lose their jobs and need to reskill and move into a new career. While Congress’ response to pandemic-related job loss demonstrated the value of a stronger safety net for displaced workers, their actions were time-limited. Programs serving dislocated workers are underfunded, provide inconsistent benefits, have complicated eligibility, and respond to cyclical rather than structural unemployment.

Policymakers can modernize these programs to make sure they meet the needs of the 21st century workforce. Workers displaced by AI will need economic supports to continue paying rent and accessing childcare as they prepare for a new career, training and education for in-demand jobs, and job retention services to thrive in new careers. In surveys of unemployed workers, these supports and services are most cited as key to preparing for, training for, and keeping a new job. AI could even be used to streamline benefit eligibility determination and administration so that it’s just as easy to access transformative re-employment services as it is to order groceries online.

Taken together, these policy actions could be a gamechanger and help level the playing field for both workers and businesses.