The impact of AI pilot zones on market competition: Causal insights from policy implementation

Research background: Artificial intelligence (AI) is a major catalyst for innovation and economic transformation, reshaping competition across industries. In 2019, China launched AI pilot zones to stimulate AI innovation while managing its economic and societal impacts. These zones aim to foster AI ecosystems through targeted investments, infrastructure, and favorable policies. However, concentration of resources and market power within these zones raises concerns about their effects on competition. This study examines how AI pilot zones influence competitive dynamics, focusing on the mechanisms driving market concentration and firm behavior. Purpose of the article: The causal impact of AI pilot zones on market competition in China is examined. A difference-in-differences (DID) methodology is used to explore the impact of these zones on firm entry, market concentration, and innovation. This study seeks to understand if pilot zones enhance competition by lowering entry barriers and supporting innovation or they exacerbate market concentration by favoring large firms. Methods: The DID methodology uses firm-level data from AI pilot and non-pilot regions (2010–2023). Market competition is assessed through concentration ratios based on revenue and income from top firms. Robustness tests, including placebo tests, and instrumental variable approaches are applied to ensure the reliability of results. Agency theory and resource dependence theory are used to explain the mechanisms through which AI policies impact competition, specifically access to resources, information asymmetry, technological diffusion and government support. Findings & value added: The study findings show that AI pilot zones positively influence competition by reducing market concentration and encouraging new firm entry. While larger firms benefit initially from policy support, the long-term effects include a more competitive market with greater firm diversification and innovation. Mechanistically, this study identifies that policy-induced enhanced information transparency, mitigated financial constraints, targeted government subsidies, and accelerated technological diffusion enable smaller firms to compete more effectively. This support helps to overcome entry barriers by providing critical resources, thereby encouraging the entry of new players and facilitating innovation. Furthermore, the policy creates a competitive environment where even incumbents must innovate to maintain their market position. The study highlights that AI pilot zones, while initially benefiting larger firms, contribute to a more competitive market by diffusing technological advancements and resources equitably across firms. This research contributes empirical evidence on the role of AI policies in shaping competition and offers valuable insights for policymakers aiming to balance innovation and competition in AI-driven economies.