Artificial intelligence readiness and innovative work behavior among Chinese university teachers: the indirect role of work engagement and the moderating role of perceived digital leadership.

INTRODUCTION: Artificial intelligence (AI) is increasingly being integrated into university teaching, research, and administration. However, it remains unclear whether and how AI readiness, as an AI-specific personal resource, is associated with broader innovative work behavior beyond AI adoption or explicitly AI-enhanced innovation. Drawing on Job Demands-Resources theory, this study examined the relationships among AI readiness, work engagement, and innovative work behavior, together with the moderating role of perceived digital leadership. METHODS: Cross-sectional survey data were collected from 432 university teachers in China. AI readiness and work engagement were modeled as reflective-reflective higher-order constructs, whereas perceived digital leadership and innovative work behavior were specified as first-order reflective constructs. The hypothesized relationships were tested using partial least squares structural equation modeling. RESULTS: AI readiness was positively associated with both work engagement and innovative work behavior, while work engagement was positively associated with innovative work behavior. The statistical indirect effect of AI readiness on innovative work behavior through work engagement was significant, with the direct association remaining significant. Perceived digital leadership strengthened the positive association between AI readiness and work engagement, although the interaction effect was very small. DISCUSSION: The findings are consistent with a resource-conversion account in which AI readiness may have cross-domain relevance for broader workplace innovation, with work engagement representing a plausible motivational pathway through which this resource is activated. Perceived digital leadership may provide a complementary enabling condition, but its limited effect warrants cautious interpretation. Given the cross-sectional, self-reported data, the results indicate statistical associations rather than causal relationships.