Measuring and Mitigating Inequality in AI-Driven Labor Markets

Abstract

The quick integration of AI into labor markets have fueled concerns of economic inequality with respect to wage polarization, job displacement, and unequitable distribution of technological dividends. While these problems are already established through various researches, lack of measures and policies for AI induced inequality is evident. This paper tries to build a comprehensive measurement system for inequality with the focus of AI influenced labor markets and policy and technology that can tackle it. It will be theorized that AI has a disproportionate benefit on high-skill labor and has made low-skill and routine labor vulnerable. Based on the mixture of methods (econometric estimation on labor markets, simulation to model potential adoption of AI) inequality metrics such as Gini coefficients, wage dispersion indices, task based measures are being considered. Policy and case study can also be adopted to compare policy and technology such as reskilling and taxation, and algorithmic fairness solutions. The findings of this study would argue that adoption of AI exacerbated current inequality, through increasing demand on high-skill and mechanization on routine labor. But well-designed policy of reskilling workforce, inclusive design of AI and redistributive policy can effectively mitigate these impacts. Early policy and action have a larger effect than reactive measures. In conclusion, this research is an effort to illustrate both the potential to worsen economic inequality and solutions by AI and contribute to the design of effective AI policy for equitable future.