Week 10: Research Conclusion and Policy Recommendations
After weeks of regression models, if there is one thing that my project has made clear, it’s that AI doesn’t affect everyone equally. Some groups—like women, older workers, certain racial minorities, and recent immigrants—are more likely to be in jobs with higher vulnerability to AI. As much as AI creates exciting new tools and careers, it also risks deepening the inequalities already existent in our labor market. But it doesn’t have to be that way – smart public policy can change the story.
The Center for Security and Emerging Technology (CSET) laid out a list of federal policy recommendations to build a more inclusive AI workforce. Here’s how some of their ideas could directly help the vulnerable groups I found in my research:
Education & Training
What I found: Workers with less education were more exposed to AI-induced job disruption.
Policy fix: CSET calls for introducing AI literacy and computer science education in K–12 schools, especially in underrepresented districts. Starting young could give everyone a fairer shot. CSET also recommends investing in non-traditional pathways to learning AI, like community colleges, certification programs, and workforce training. These can help people without a four-year degree pivot into more stable jobs.
Income-Based Inequality
What I found: Higher-income workers actually face more AI exposure because they’re in jobs that involve data, analysis, and routine digital tasks.
Policy fix: CSET suggests upskilling opportunities such as short-term AI courses or on-the-job training to help people adapt instead of being displaced.
Support for Immigrants
What I found: Recent immigrants start in more exposed jobs.
Policy fix: Targeted support—like language-inclusive training programs and easier access to certifications—can help speed up this transition and reduce vulnerability.