How artificial intelligence-induced job insecurity shapes knowledge dynamics: the mitigating role of artificial intelligence self-efficacy

This research examines the intricate relationships between artificial intelligence (AI)-induced job insecurity, psychological safety, knowledge-hiding behavior, and self-efficacy in AI learning within organizational contexts. As AI technologies increasingly permeate the workplace, comprehending their impact on employee behavior and organizational dynamics becomes crucial. Based on several theories, we use a time-lagged research design to propose and test a moderated mediation model. We collected data from 402 employees across various industries in South Korea at three different time points. Our findings reveal that AI-induced job insecurity positively relates to knowledge-hiding behavior, directly and indirectly, via reduced psychological safety. Moreover, we discover that self-efficacy in AI learning moderates the relationship between AI-induced job insecurity and psychological safety, such that high self-efficacy buffers the harmful influence of job insecurity on psychological safety. These results enhance the existing literature on organizational technological change by clarifying the psychological processes through which AI implementation influences employee behavior. Our study highlights the critical role of psychological safety as a mediator and self-efficacy as a moderator in this process. These insights present significant implications for managers and organizations navigating the challenges of AI integration. They emphasize the need for strategies that foster psychological safety and enhance members’ confidence in their ability to adapt to AI technologies. Our research underscores the significance of considering both the technical and human aspects of AI implementation within organizational contexts.