Artificial intelligence, corporate information governance and ESG performance: Quasi-experimental evidence from China

Abstract

Artificial intelligence (AI) has emerged as a transformative force in global business, yet its impact on environmental, social, and governance (ESG) remains largely unexplored. This study examines how AI contributes to enhancing corporate ESG outcomes, using China's national AI pilot zones as a quasi-natural experiment. By applying a staggered difference-in-differences approach to panel data from 1418 A-share listed firms in China (2011−2022), we find that AI significantly improves ESG performance. The primary mechanism behind this improvement is better corporate information governance, which includes improved information transparency, enhanced information sharing, and reduced information asymmetry. These improvements, in turn, lead to improved environmental information disclosures, optimized supply chain management, and reduced agency costs. Private firms, as well as firms with lower institutional attention and higher equity concentration, experience a greater positive effect. Our findings shed light on the mechanisms through which AI influences corporate ESG performance, emphasizing the potential of AI to strengthen corporate governance and support sustainability efforts.