The impact of artificial intelligence on global energy vulnerability

Abstract

Investigating the effect of artificial intelligence (AI) on energy vulnerability (EVI) is crucial to understanding how technological advances are changing the resilience and sustainability of energy systems. However, their quantitative relationship still lacks empirical evidence. This study first constructs the EVI of 54 global economies from the perspective of energy security, energy consumption, energy efficiency, and energy availability from 2000 to 2019. Then, a fixed-effect model is employed to investigate the relationship between AI and EVI. Results show that (1) AI can considerably reduce global EVI. The core findings remain reliable after several robustness checks. (2) Mechanism analysis implies that AI can reduce EVI by promoting financial development and technological progress. (3) Heterogeneity analysis implies that the impeding role of AI on EVI is more pronounced in countries with low incomes and industrialization levels. Furthermore, the hindering effect of AI on EVI is strengthened after Industry 4.0 and the financial crisis. Some policy implications are further proposed accordingly to reduce global EVI.