How does artificial intelligence usage affect the safety behavior of bus drivers? A double-edged sword study

This study aims to provide a framework for understanding the mechanism by which artificial intelligence (AI) usage affects the safety behavior of bus drivers through cognitive appraisal theory. We examined data from 555 bus drivers at three-time points. Our findings indicate that AI usage is positively related to both safety self-efficacy and job insecurity, which in turn are linked to safety behavior. Safety self-efficacy and job insecurity mediate the relationship between AI usage and safety behavior. Additionally, we found that trait resilience moderates the positive relationship between AI usage and safety self-efficacy, as well as the relationship between AI usage and job insecurity. Furthermore, trait resilience moderates the indirect effect of AI usage on safety behavior through safety self-efficacy and job insecurity. The results suggest that AI usage has two faces, both enhancing and impairing the safety behavior of bus drivers. These findings are crucial for management theory and practice.