AI Awareness and Employee Innovation: A Dual-Pathway Moderated Mediation Model Within Organizational Systems

Abstract

Amid AI-driven organizational transformation, employees’ subjective evaluations of AI technologies—referred to as AI awareness—have become a critical psychological factor influencing innovation behavior. This study aims to uncover how AI awareness affects employee innovation performance through behavioral pathways and examines the moderating role of individual achievement motivation in this process. Grounded in Cognitive Appraisal Theory and the Dual Pathway Model, we construct a dual-path mediation model, in which proactive and withdrawal behaviors act as mediators, and achievement motivation serves as a boundary condition. Based on a two-wave survey of 413 knowledge workers in China’s high-tech sector, the proposed model was empirically tested using moderated mediation analysis (PROCESS macro). The results reveal that AI awareness has a significant dual-path effect on innovation behavior: on the one hand, it promotes innovation by stimulating proactive behavior; on the other hand, it may suppress innovation by inducing withdrawal behavior. Moreover, achievement motivation plays a crucial moderating role in this mechanism by strengthening the positive mediating effect of proactive behavior and weakening the negative mediating effect of withdrawal behavior. This study conceptualizes AI awareness as a psychological input encompassing both perceived opportunity and perceived threat, clarifies the behavioral response mechanisms of employees facing AI, and highlights the boundary-regulating role of individual motivation in organizational adaptability. Practically, the study suggests designing differentiated HR interventions based on employees’ cognitive appraisals and motivational profiles to enhance human–AI collaboration, foster innovation resilience, and improve organizational adaptability amid digital transformation.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 27 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.