Not yet human-reviewed
This source record was written automatically by the research pipeline and has not yet been reviewed by the maintainer or integrated into a topic page. Treat it as a provisional summary and verify claims against the original source.
Automation and Augmentation: Artificial Intelligence, Robots, and Work
Citation. Lei, Y.-W., & Kim, R. (2024). Automation and augmentation: Artificial intelligence, robots, and work. Annual Review of Sociology, 50, 251–272. https://doi.org/10.1146/annurev-soc-090523-050708
Summary. This peer-reviewed review synthesizes sociological and adjacent research on robots and AI in task automation and labor augmentation. It argues that displacement remains consequential but is shaped by task diversity, dexterity, flexibility, tacit knowledge, collective institutions, and policy; automation and augmentation can also coexist within economic change. The review treats AI outcomes as socially contingent rather than technologically determined, emphasizing workplace control, autonomy, inequality, and the organizational and political conditions that shape whether augmentation is beneficial.
Key claims
- Task automation continues to produce displacement effects, but jobs involving diverse tasks, physical dexterity, flexibility, tacit knowledge, or protection through professional associations or unions are less likely to be negatively affected.
- Automation and augmentation are not mutually exclusive: technological change can replace human input in some tasks while enabling or creating other work, so aggregate employment and productivity outcomes depend on countervailing effects rather than a single technological direction.
- Without policy intervention, automation and augmentation can widen inequality across social groups, labor and capital, and firms because better-resourced actors are better positioned to avoid harms and capture gains.
- AI-enabled workplace systems can produce dissatisfaction and disengagement when they are opaque, error-prone, insensitive to context or tacit knowledge, impose extra work, or leave workers without meaningful authority to override assessments.
- Research on AI at work should identify the social and political conditions and mechanisms that reduce harms and support labor augmentation rather than treating technical capability as a sufficient explanation of outcomes.
Evidence & limitations
- Narrative review article spanning economic, sociological, and related literature on automation, robots, and AI; it synthesizes existing findings rather than reporting a new causal study or meta-analytic effect estimate.
- Its broad, multi-level framing supports contingent claims about work and inequality, but individual empirical findings and causal mechanisms should be checked against the primary studies it cites.
- The article is peer-reviewed and published open access under a CC BY 4.0 license in Annual Review of Sociology.