Abstract Fears of technological unemployment often portray automation as a force that eliminates occupations. This study offers a different perspective by modelling automation as the sequential erosion of tasks, which reshapes occupational skill bundles and mobility structures. Using data from the Occupational Information Network (O*NET), integrated with two exposure measures—routine task automation and AI-driven cognitive automation—we simulate how the removal of 332 tasks alters skill requirements across 736 occupations. Results suggest that automation increases skill overlap between occupations, promoting structural integration within the occupational network. Yet the nature of integration diverges: routine automation primarily dismantles specialised physical skills, enhancing mobility only within homogeneous manual clusters, whereas AI automation moderates a broader range of cognitive and social skills, creating new bridges across heterogeneous domains. Despite substantial task erosion, most occupations retain residual skills that enable adaptation rather than extinction. By tracing changes in the shares of skills reallocated to machines, we explore how AI-driven automation sustains occupational roles through emerging complementarity rather than substitution. While the model is not designed to forecast labour market outcomes or to conduct counterfactual tests, the results theoretically reframe automation as a process of reorganisation that may expand, rather than constrain, labour mobility. Policy responses must therefore move beyond predicting job loss to supporting workers in navigating newly emerging, and often counterintuitive, mobility pathways.
Lee et al. (Tue,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: