How Human-Centered AI Buffers the Negative Effects of AI-Induced Job Insecurity on Sustainable Innovation

초록

The integration of artificial intelligence (AI) generates employee concerns regarding job displacement, termed AI-induced job insecurity (AIJI). This study examines how AIJI influences sustainable innovation through the serial mediation of psychological safety and knowledge-sharing behavior, and whether Human-Centered AI (HCAI) buffers these effects. Analyzing multi-wave data from 407 employee-HR dyads across Korean industries, we found that AIJI does not directly suppress sustainable innovation. Instead, it operates through a sociopsychological pathway: structural threats erode psychological safety, which inhibits knowledge sharing, ultimately hampering sustainable innovation. Crucially, HCAI acts as a buffer; when AI transparency and user agency are high, the negative impact of insecurity on psychological safety is neutralized. These findings advance conservation of resources and threat-rigidity theories, suggesting that realizing AI's potential requires prioritizing human-centric design to preserve the social fabric essential for collective creativity.

키워드