Artificial Intelligence and Energy Resilience: Dynamic Evolution, Developmental Heterogeneity, and the Moderating Roles of Geopolitical Risk and Clean Energy

ABSTRACT Enhancing energy resilience contributes to ensuring energy security and stable operation, and artificial intelligence (AI) offers a new technological path. This study explores the relationship between AI and energy resilience using data for 42 countries from 2010 to 2022. First, based on five dimensions—total energy consumption, energy access, energy efficiency, renewable energy, and energy security—this study constructs a comprehensive evaluation system for energy resilience. The projection pursuit model based on a genetic algorithm is applied to measure resilience levels, revealing the nonlinear relationship between AI and energy resilience. A quadratic moderating effect model and curve simulation techniques are used to analyze the mechanisms of external risks and clean energy impact. The study shows that AI initially enhances energy resilience through prediction, fault diagnosis, and intelligent decision‐making. However, technological risks such as data breaches, communication disruptions, cyber threats and attacks weaken the resilience‐enhancing effect. Compared with developing countries, developed countries exhibit a more significant effect of AI technology on improving energy resilience. Furthermore, rising geopolitical risks will reshape the effects of AI technology, and the synergistic effect of AI and clean energy can further strengthen its positive effect on energy resilience. This study provides new insights into optimizing the application of AI, contributing to the construction of a safer and more stable energy system.