Answers are grounded in this paper and cite their sources.
This study examines the relationship between artificial intelligence (AI) labor intensity and workplace safety violations in the architecture, engineering, and construction (AEC) industry where foundational AI technologies have been incorporated into practice for decades before the emergence of generative AI. This paper hypothesizes that managers overrely on AI-enabled monitoring systems and consequently pay less attention to traditional safety-management practices. Using a novel firm-level data set of AI worker share as a proxy of AI labor intensity, it finds that firms with greater AI worker share experience significantly more workplace safety violations. The effect is stronger in AEC firms than in nonAEC firms and is concentrated in serious violations. Additional analyses show more victims, larger regulatory penalties, and weaker safety environments for AI-intensive AEC firms. However, strong board monitoring mitigates the adverse relationship between AI worker share and workplace safety violations. Overall, these findings suggest that the benefits of early AI adoption depend on complementary governance structures and sustained managerial attention to workplace safety.
No takes yet. Share an insight, caveat, or question.
Eunpyo Hong (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: