Researchers have developed an artificial intelligence system capable of detecting landslide risks near electricity transmission towers, offering a proactive solution to prevent infrastructure failures. The breakthrough, detailed in the International Journal of Power and Energy Conversion, leverages remote-sensing change detection to identify geological disturbances that traditionally challenge automated monitoring systems due to limited historical disaster datasets. The new model employs a twin-network architecture that processes and compares satellite or drone imagery captured before and after environmental events. To enhance accuracy, the system integrates a visual foundation model pretrained on extensive remote-sensing data, providing comprehensive terrain and landscape context. A critical component is the attention-based alignment module, which filters out non-hazardous variables such as seasonal vegetation shifts and lighting variations. By isolating these factors, the AI selectively highlights structural ground changes directly linked to geological instability. Field testing against established change-detection benchmarks demonstrates that the framework consistently outperforms recent methodologies. This advancement enables utility operators to anticipate slope failures and geological degradation before they compromise critical power transmission assets. By transforming infrequent and fragmented disaster imagery into actionable geospatial intelligence, the technology establishes a reliable early-warning mechanism for energy grid resilience. The system marks a significant step toward automating infrastructure risk assessment, reducing unplanned outages, and minimizing costly emergency repairs in geologically vulnerable regions.