Impact of AI on corporate ESG performance: Evidence from China

Artificial intelligence (AI) technology plays a crucial role in achieving sustainable development, yet its impact on corporate ESG performance remains underexplored with inconsistent findings. Based on data from Chinese A-share listed companies in Shanghai and Shenzhen between 2010 and 2023, this paper employs the large language model ERNIE 3.0 to conduct text analysis of the management discussion and analysis sections of annual reports, constructs an AI technology adoption index for enterprises, and investigates the impact of AI technology adoption on corporate ESG performance and its underlying mechanisms. The study found that AI technology adoption significantly improved corporate ESG performance, with this positive effect being more pronounced in the sub-indicators of the social (S) dimension. Following a series of robustness tests, the empirical results of this paper remain robust. The primary impact mechanism is reflected in four channels through which AI technology adoption influences corporate ESG performance: enhancing information transparency, alleviating financing constraints, promoting green technology innovation, and increasing total factor productivity. The results of the heterogeneity analysis indicate that AI technology adoption has a significant positive impact on the ESG performance of non-state-owned enterprises, non-high-tech enterprises, and labor-intensive enterprises. Compared to the western region, AI technology adoption in the eastern and central regions has a more pronounced effect on improving ESG performance. This study highlights the significance of AI technology adoption in enhancing corporate ESG performance. The research findings provide empirical support and policy implications for promoting the development of AI technology in enterprises and fostering a greener, more sustainable economic system.