Digital decoupling: educational stratification and dual-track effects of AI displacement and augmentation in U.S. occupations

This study investigates the phenomenon of digital decoupling by proposing an AI Dual-Track model designed to quantify how generative artificial intelligence (AI) simultaneously displaces and augments labor across 846 U.S. occupations. By leveraging an ensemble-based AI classification protocol, the research constructs a validated set of 63 O*NET competencies to map occupational tasks into substitution and facilitation tracks. Central to this analysis is the exploration of how educational attainment stratifies these dual-track dynamics. The findings reveal an exposure paradox: although AI-driven displacement permeates the full occupational distribution, augmentation gains and associated economic returns remain disproportionately concentrated in occupations requiring higher levels of formal education. This pattern points to a form of digital decoupling, in which the central divide is not access to AI itself, but the capacity to leverage AI in ways that expand rather than merely replace human labor. Consequently, institutional credentials serve as a critical mediator for the facilitation track, determining a worker’s ability to translate AI exposure into productivity gains. The resulting data indicate that over 9.1 million worker equivalents in middle-skill occupations face significant displacement pressures, while higher education attainment groups capture the largest gains in net economic capacity. These results suggest that the effects of AI on work are not technologically predetermined but instead are structured through differential access to augmentation-enabled competencies. The study concludes by highlighting the policy importance of expanding access to AI facilitation capabilities, framing facilitation literacy as a potential public capability necessary for more inclusive participation in an AI-mediated labor market.