This session explores the integration of geospatial data and artificial intelligence techniques for comprehensive multi-hazard analysis and disaster management. With the increasing availability of remote sensing data and advances in machine learning and deep learning, Geospatial AI (GeoAI) offers powerful tools to monitor, model, and predict diverse natural hazards, including floods, earthquakes, landslides, and wildfires. The session will highlight innovative methodologies for spatiotemporal analysis, data fusion, and risk assessment, as well as approaches for early warning systems and decision support. Emphasis will be placed on explainable and trustworthy AI models to enhance transparency and operational reliability in critical applications. Contributions addressing real-world case studies, large-scale implementations, and interdisciplinary approaches are particularly encouraged. This session aims to foster collaboration to advance resilient and data-driven disaster management strategies in a changing global environment.