The impact of artificial intelligence development on embodied carbon emissions: Perspectives from the production and consumption sides

Abstract

The positive impact of artificial intelligence (AI) on carbon emission issues has been widely discussed, but no research has yet thoroughly examined the specific effects of AI on embedded carbon emissions (ECE). To address this gap, this study examines the impacts and underlying mechanisms of AI development on ECE from both production and consumption perspectives. The results show that AI development significantly reduces carbon emissions on both fronts, with a more noticeable effect on the production side. Specifically, AI has generated significant carbon reduction effects on ECE from both the production and consumption sides by optimizing manufacturing structures, promoting low-carbon energy consumption, and driving green product innovation. Additionally, factors such as strengthened environmental regulations, improved ecological civilization, improved green innovation capacity, and higher renewable energy consumption further strengthen the carbon reduction effects of AI. However, the impact of AI on carbon emissions shows regional heterogeneity. This study offers a new perspective on differentiated carbon governance and equitable allocation of carbon emission responsibilities. It also offers empirical evidence and theoretical insights for policymakers, supporting the effective use of AI technology to achieve carbon reduction goals.