Natural hazards and extreme events pose major challenges for monitoring and risk management across space and time, particularly due to interconnected exposure, vulnerability, and cascading impacts. During rapidly evolving crises, timely and reliable information on hazard evolution, exposure, damage, and impacts is essential. Before disasters occur, up-to-date information on exposure, vulnerability, critical infrastructure, and the built environment is equally important for preparedness, mitigation, and establishing reliable baseline conditions. Advances in Earth observation (EO), UAVs, ground sensors, IoT networks, and geospatial infrastructures are enabling near-real-time monitoring from local to global scales. In parallel, artificial intelligence (AI) and GeoAI are transforming the analysis of large and heterogeneous geospatial datasets, supporting hazard detection and forecasting, susceptibility and risk assessment, damage mapping, and decision support. They also provide new opportunities for continuously updating exposure and vulnerability information. Despite this progress, challenges remain in moving from experimental models to reliable operational systems, including model generalization, data scarcity and imbalance, multi-sensor fusion, uncertainty, explainability, scalability, reproducibility, and trustworthy AI. Interoperability with disaster-management standards and operational workflows is also essential for effective uptake by civil protection and emergency-management authorities. This session welcomes methodological, experimental, and operational studies combining remote sensing, geospatial information, and AI for floods, wildfires, landslides, earthquakes, volcanic hazards, droughts, coastal hazards, extreme weather, land deformation, subsidence, and multi-hazard or cascading events. Topics include rapid hazard mapping and forecasting; susceptibility, exposure, vulnerability, and risk assessment; damage analysis; explainable, trustworthy, and physics-informed AI; satellite-UAV-ground data fusion; transfer learning and foundation models; decision-support systems; and integration of EO/AI products into disaster-management practice.We particularly encourage studies that bridge research and operations, demonstrate transferability across regions or hazard types, provide open datasets or reproducible workflows, and connect scientific information products with established disaster-management practices.