Week 3: Why These Demographics Matter in AI Job Displacement
Hey everyone, welcome back to my blog! This week, I’ve been diving deeper into why we are analyzing specific demographic variables.
Why Are We Looking at These Variables?
To understand AI-driven job displacement, we need to analyze how different demographic groups are affected. The variables I selected—sex, education, race, job type, income, and duration of residency in the U.S.—are crucial because historical data shows that technological advancements have not impacted these demographics equally. Below are some of the variables I am analyzing:
Sex
Gender disparities in the workforce are well-documented. While automation has, at times, benefited women—such as the shift from physically intensive labor to service-oriented jobs, which brought more women into the workforce during the nineteenth and twentieth centuries—it has also reinforced inequalities. Research shows that today, women are overrepresented in administrative and customer service roles, both of which are highly exposed to AI automation. By examining the relationship between sex and exposure to AI, I aim to determine whether AI-driven job losses disproportionately affect women or men.
Education
Education level is a strong predictor of job security in an AI-driven economy. Some studies (such as one from the Pew Research Center) suggest that jobs requiring only a high school diploma are at a much lower risk of automation than those requiring a bachelor’s degree or higher. This is because many blue-collar and manual labor jobs—such as barbers, dishwashers, and passenger assistants—rely on physical skills and human interaction, making them less easily automated. In contrast, white-collar jobs requiring advanced degrees often involve repetitive cognitive tasks, such as data analysis or legal document review, which AI systems can efficiently handle. By including education as a variable, we can determine whether AI is in fact exacerbating job losses for workers with higher levels of education.
Race
Historically, racial disparities have played a significant role in employment trends. Researchers strongly disagree on which demographics are most vulnerable to AI, with some arguing that Asian and White workers are more likely to be employed in roles with high automation potential, while others claim Blacks and Hispanics to be most vulnerable. According to a study in 2024, certain racial minorities also face greater barriers to reskilling opportunities. Including race as a variable in our analysis helps us assess whether AI is disproportionately affecting certain racial groups and deepening existing inequalities.
Job Type
Different job types have varying levels of exposure to AI automation. White-collar roles, particularly those in administrative, analytical, and customer service fields, face higher automation risks due to advances in natural language processing and machine learning. In contrast, jobs that require creativity, emotional intelligence, or hands-on labor are less susceptible to automation. Understanding how job type influences AI exposure will be crucial in predicting future employment trends.
What’s next?
In the coming weeks, I will be running polynomial regressions, using the demographic as the independent and AI Exposure as the dependent variable. However, in order to do this, I first need to find the optimal degree for my data, to minimize over- and underfitting. I will also be researching Occupational Exposure to AI indexes established by past studies to use for my research.
Thanks for tuning in, and I will keep you updated on my research in the coming weeks!
Comments:
All viewpoints are welcome but profane, threatening, disrespectful, or harassing comments will not be tolerated and are subject to moderation up to, and including, full deletion.