Analysis of the impact of digital economy on carbon emissions and its mediation effect

Against the backdrop of urban green low-carbon transformation and AI-empowered digital economy development, examining the impact of digital economy on carbon emissions has become a crucial research topic. This study employs the entropy weight method to calculate the digital economy development index, utilizing mediation effect models and threshold regression models to empirically analyze the impact and mechanisms of digital economy on carbon emissions. The results reveal that: (1) An inverted U-shaped relationship exists between digital economy development and carbon emissions, with initial increases followed by decreases after reaching a certain value; (2) Electricity consumption and green technology innovation serve as mediation mechanisms; (3) Digital economy can effectively curb emissions when urbanization and financial development surpass certain threshold levels. Based on these research findings, this study proposes targeted policy recommendations that make substantive contributions to low-carbon sustainable development.