The role of artificial intelligence in enhancing corporate environmental information disclosure: Implications for energy transition and sustainable development

Publication Name: Energy Economics

Publication Date: 2025-08-01

Volume: 148

Issue: Unknown

Page Range: Unknown

Description:

Global climate and environmental issues pose severe challenges to the sustainable development of human society. As major contributors to environmental pollution and carbon emissions, the quality of enterprises' environmental data has gained significant attention in academic and industrial circles. This study analyzes information from Chinese A-share companies spanning 2012 to 2023 to investigate the pathways through which artificial intelligence (AI) technology influences corporate environmental information disclosure (EID). The results indicate that AI significantly enhances the quality of corporate EID by optimising internal control levels and strengthening external supervision mechanisms. These conclusions have been validated through robustness and endogeneity tests. The heterogeneity analysis further reveals that the promoting effect of AI is more significant in large corporates, corporates in central cities, mature corporates, corporates audited by the Big Four international accounting firms, high-tech corporates, and heavily polluting industries. The study innovatively constructs a dual-path theoretical framework of ‘internal management optimisation–external supervision strengthening’ and integrates macro urban AI indicators with micro enterprise data, contributing new empirical support for the digital transformation and green governance of developing countries. Based on these findings, policymakers should promote the innovative application of AI technology in corporate environmental governance, improving internal control norms, optimising the external supervision system, and implementing a classified guidance strategy for different enterprise attributes, so as to help enterprises achieve low-carbon transformation and sustainable development.

Open Access: Yes

DOI: 10.1016/j.eneco.2025.108680

Authors - 4