Lu Yao, Liao ZeFang
College of Economics and Management, Shanghai Ocean University, No. 999, Hu Cheng Ring Road, Shanghai, 201306, China.
Sci Rep. 2025 Apr 12;15(1):12585. doi: 10.1038/s41598-025-97110-3.
As a critical aspect of the industry 4.0 era, the application of artificial intelligence (AI) is significant to environmental governance. It serves as a crucial driving force in assisting enterprises in the transition toward low-carbon practices. This paper examines China's A-share industrial enterprises from 2011 to 2022, constructs and trains a word vector model to extract AI-related terms, and the impact of AI applications on the carbon emission intensity of these enterprises is investigated. The findings reveal that enhancing the level of AI application can effectively decrease carbon emission intensity. Specifically, a 1% increase in AI application leads to a reduction of 0.0395% in carbon emission intensity. Further analysis indicates that enterprises can diminish their carbon emission intensity by the optimization of supply chain and green technology innovation. Heterogeneity analysis suggests that utilizing AI is beneficial for reducing the carbon emission intensity of manufacturing, high-tech, and high-pollution enterprises. The results of this study enrich the micro-level research on the relationship between AI and carbon emission intensity, offering valuable insights for enterprises aiming to achieve sustainable development.
作为工业4.0时代的一个关键方面,人工智能(AI)的应用对环境治理具有重要意义。它是协助企业向低碳实践转型的关键驱动力。本文考察了2011年至2022年中国A股工业企业,构建并训练词向量模型以提取与人工智能相关的术语,并研究了人工智能应用对这些企业碳排放强度的影响。研究结果表明,提高人工智能应用水平可以有效降低碳排放强度。具体而言,人工智能应用水平每提高1%,碳排放强度就会降低0.0395%。进一步分析表明,企业可以通过优化供应链和绿色技术创新来降低其碳排放强度。异质性分析表明,利用人工智能有利于降低制造业、高科技企业和高污染企业的碳排放强度。本研究结果丰富了关于人工智能与碳排放强度关系的微观层面研究,为旨在实现可持续发展的企业提供了有价值的见解。