Generative artificial intelligence (GenAI) is rapidly transforming organizational knowledge processes, yet little is known about how its knowledge-generative capacity translates into innovation outcomes. This study aims to examine how GenAI-augmented knowledge management (GAIKM) enhances innovation performance through the activation of dynamic capabilities and investigates how learning orientation conditions this process.
Drawing on knowledge-based view and process-oriented knowledge management theory, this study conceptualizes GAIKM through two distinct yet complementary processes: GenAI-augmented knowledge creation (GAIKC) and GenAI-augmented knowledge integration (GAIKI). A theory-driven model is tested using survey data from 224 technology-based firms in China. Partial least squares structural equation modeling is used to test the hypotheses, supported by reliability, validity, mediation and moderation analyses.
Results show that both GAIKC and GAIKI strengthen sensing, seizing and transforming capabilities, with GAIKI exhibiting stronger overall effects. Dynamic capabilities mediate the relationship between GAIKM and innovation performance, with seizing capability producing the strongest indirect effect in this context. Learning orientation enhances the effects of GAIKC on sensing capability and GAIKI on seizing capability but weakens the effects of GAIKC on transforming capability.
This study conceptualizes GAIKM as a dual-process knowledge activation mechanism and extends dynamic capability theory by explicating how AI-enabled knowledge processes translate into innovation through differentiated capability pathways. It further highlights learning orientation as a boundary condition shaping the transformation of GenAI-generated knowledge into organizational capabilities and provides practical guidance for managers to build GenAI-based knowledge systems and learning cultures that enhance sustainable innovation performance.
