Preparing workers for AI: new report highlights policy gaps and solutions

A new report takes a wide-ranging look at how the rapid rise of artificial intelligence introduces both opportunity and risk for workers. Titled “Proactively Developing and Assisting the Workforce in the Age of AI,” the paper grew out of a Washington, D.C., workshop and delves into how AI is impacting labor markets. It identifies gaps in policy and outlines options for responding to AI-driven disruptions to help broaden economic opportunity.  

Upjohn Institute President Mike Horrigan was a major contributor, authoring Chapters 4, 5, and 6 of the report, which is published by Americans for Responsible Innovation and the University of Notre Dame’s Keough School of Global Affairs. 

The paper looks at policy options covering four areas: 

It concludes that to navigate AI’s upheaval of the job market, policymakers must embrace proactive, evidence-based strategies that foster agility, equity, and adaptation.    

Chapter 4: The U.S. labor policy landscape 

Horrigan and his co-authors write that labor and education policies are foundational for equipping both workers and institutions with the skills they will need in a labor market altered by AI. These policies must grow workforce skills, align those skills with employer needs, retrain workers after job dislocations, and provide access to education and training. 

As the impact of AI increases, spotting the merits and gaps in these policies is critical to helping the workforce land on its feet. To assist in this evaluation, Horrigan organizes policies and programs by educational type: 1) K–12 education, 2) career and technical education, 3) higher education, 4) other workforce programs and institutions, and 5) work-based learning. He then takes an in-depth look at each category and discusses how it can employ AI to foster workforce development.  

Chapter 5: Data, research, and measurement  

The authors propose ways to anticipate the rapidly changing and unpredictable landscape of AI. They advocate modernizing the federal statistical system, improving real-time data collection, and encouraging robust collaboration among institutions to track AI’s evolution. 

These improvements would enable better forecasting and precision interventions, allowing policymakers to be agile and responsive to shifts driven by AI. Government data systems need to be timely, granular, and flexible. Horrigan, who in his former capacity as associate commissioner of the U.S. Bureau of Labor Statistics oversaw the nation’s employment, unemployment, and inflation measurement programs, outlines several strategies, including real-time adaptive mapping of the latest occupations, tasks, and skills needed to keep pace with the rapidly changing nature of work from AI. 

Chapter 6: Workforce development and education 

Horrigan and co-authors caution that AI is advancing so rapidly it has overtaken the efforts of organizations to prepare their workers for it. One study found that 64 percent of workers worry they will be left behind if they don’t assimilate to using AI. Yet 82 percent say their organizations have failed to offer training on how to use AI tools.   

Readying the workforce for AI requires a broad, human-centered approach emphasizing critical thinking, empathy, adaptability, and collaboration. It may also mean encouraging workers to reimagine their career pathway, fostering a culture of lifelong learning, and revamping education systems.   

These proposals and others covered in the paper, promising as they are, should be rigorously evaluated for cost-effectiveness. Over time, knowledge of what works and what doesn’t can direct our national resources to tested policies and practices. 

The workshop on AI and labor policy was convened by the Keough School of Global Affairs and the Institute for Ethics and the Common Good at the University of Notre Dame, and held in Washington, D.C., on March 11, 2025.