Artificial Intelligence Adoption and Innovation Outcomes: Mechanisms and Empirical Evidence

Abstract

The integration of Artificial Intelligence (AI) technologies has become a key driver of enterprise innovation in the digital economy. However, the impact of AI adoption on the innovation quality (INQ) of firms remains underexplored. This study investigates how AI adoption influences INQ and reveals the underlying mechanisms that mediate this relationship. Using panel data from Shanghai and Shenzhen A-share listed firms from 2010 to 2021, the paper employs fixed-effect and robustness regression models to empirically analyze the effects of AI adoption on INQ. The results show that enterprise AI adoption significantly enhances INQ at the 1% confidence level, and this positive relationship remains robust across multiple model specifications and subsample tests. Mechanism analysis further demonstrates that innovation cooperation (IC) plays a mediating role in this relationship, suggesting that AI adoption promotes collaborative innovation and thus improves INQ. Moreover, heterogeneity analysis reveals that the positive effect of AI on INQ is more pronounced among state-owned enterprises, firms located in eastern regions, and those operating in regions with stronger intellectual property protection (IPP). These findings provide new empirical evidence that AI technologies not only improve innovation efficiency but also elevate innovation quality by enhancing cooperation, knowledge sharing, and resource integration. The study offers valuable implications for policymakers and managers seeking to leverage AI for sustainable innovation-driven development.