Strategic adoption of generative AI in organizations: A game-theoretic and network-based approach

Abstract

The rapid advancement of Generative AI (GenAI) has introduced new opportunities and challenges for organizations seeking to integrate AI-driven decision-making into hierarchical structures. This study presents a game-theoretic framework combined with a hybrid organizational network model to examine the dynamics of GenAI adoption. The network consists of managers, employees, and AI systems interacting within a structure that combines hierarchical reporting and scale-free collaboration. By simulating multiple organizational scenarios with varying adoption costs, managerial influence, and network rewiring mechanisms, we analyze how AI adoption propagates through departments, impacts interdepartmental collaboration, and influences organizational inequality. Our findings suggest that managerial optimism and the ability of organizational networks to adapt flexibly accelerate AI adoption. On the other hand, decentralized decision-making enhances collaboration but may lead to short-term inefficiencies. Furthermore, although adopting AI initially leads to increased inequality within the organization, this disparity tends to stabilize over time. Therefore, effective governance strategies are critical in balancing organizational efficiency with fairness. This research provides actionable insights for managers and policymakers to navigate AI integration effectively and optimize its long-term benefits.