Knowledge sharing and organizational adaptability through workforce dynamic capabilities and Human-AI collaborative tasks

Purpose This study aims to investigate why firms often remain slow to adapt to artificial intelligence (AI)-led change despite substantial investment in learning systems and knowledge sharing. It develops a clearer explanation of organizational adaptability by examining the roles of AI-related knowledge sharing, coordinated workforce actions and boundary-spanning leadership in AI-intensive work settings. Design/methodology/approach This study adopts a theory-driven empirical approach grounded in the knowledge-based view. The proposed model is tested using survey data from 485 validated responses collected from Chinese enterprises operating across a wide range of industries at both national and international levels. Structural equation modelling is used to estimate the main relationships, while the Gaussian copula approach addresses potential endogeneity concerns. The findings are further supported through agent-based simulation and sensitivity analyses, which confirm the robustness of the results. Findings The findings indicate that AI knowledge sharing is positively associated with organizational adaptability. Workforce dynamic capabilities partly explain this relationship, and the association is stronger when boundary-spanning leadership is present. In addition, the link between workforce dynamic capabilities and organizational adaptability is partly explained by human–AI collaborative tasks, suggesting that AI knowledge sharing supports adaptability by strengthening adaptive capacity and by enabling its expression in AI-supported work. Originality/value The study’s originality lies in providing a more precise theoretical explanation of how firms adapt effectively to AI-led change. It advances a dynamic enactment perspective within the knowledge-based view by showing that AI-related knowledge creates adaptive value when it is continuously interpreted and enacted in everyday work practices.