Research on How AI Substitution and Complementarity Effects Shape Labor Market Polarization

Abstract

This study investigates how artificial intelligence reshapes labor market structures through two distinct mechanisms: substitution, where AI replaces human tasks, and complementarity, where AI augments human capabilities. Drawing on granular occupational task data and longitudinal employment trends, we identify a non-uniform pattern of labor market polarization—one that diverges from classical skill-biased technological change models. Rather than a simple hollowing out of middle-skill jobs, we observe asymmetric displacement in routine cognitive roles alongside accelerated upgrading in interactive, adaptive, and context-sensitive occupations. Complementarity effects are strongest where human judgment, ethical reasoning, and real-time socio-emotional coordination remain indispensable. The findings challenge monolithic narratives about AI-driven job loss and reveal a more nuanced restructuring process, wherein occupational boundaries blur, skill demands reconfigure, and labor value migrates toward uniquely human capacities that resist algorithmic replication.