Abstract: Based on a balanced panel dataset covering 31 provincial-level administrative regions in China from 2011 to 2023, this study employs a two-way fixed effects model and a Spatial Durbin Model, combined with mediation and moderation tests, to systematically identify the impact, mechanisms, and variations of artificial intelligence (AI) on provincial economic resilience. The findings show that AI development can significantly enhance the level of economic resilience, and this conclusion remains robust after replacing core variables, adjusting the sample scope, and addressing potential endogeneity. The influence of AI on economic resilience is not a single linear effect; rather, it manifests simultaneously through mechanism transmission, conditional constraints, and regional heterogeneity. Economic resilience exhibits notable spatial autocorrelation, and AI development demonstrates significant spatial spillover effects. Provincial economic management authorities should continue to promote AI development and consolidate its technological foundation for empowering economic resilience; refine supporting policies around transmission mechanisms and operating conditions to improve the conversion efficiency of AI into resilience gains; and strengthen regional coordination and spatial linkage to facilitate the effective release of AI’s spillover effects in empowering economic resilience.

Key words: artificial intelligence, economic resilience, spatial spillover effect, mediation effect