In order to examine the relationship between new infrastructure investment and carbon emission intensity, provincial panel data from 2010 to 2020 is used to construct a two-way fixed-effects model for empirical analysis of its impact and mechanism. The findings indicate that new infrastructure investment exerts a significant inhibitory effect on carbon emission intensity, with energy consumption intensity playing a positive moderating role in the process. Furthermore, the effects of new infrastructure investment on carbon emission intensity exhibit regional heterogeneity, with significant inhibitory effects observed in both eastern and central regions. Moreover, these inhibitory effects are more pronounced in high-integrated computing regions.