Our understanding of volcanic hazards is evolving rapidly, driven by breakthroughs in satellite Earth observation, novel ground-based instruments, and artificial intelligence. The integration of artificial intelligence techniques, including machine learning, facilitates the rapid analysis of vast datasets, uncovering hidden patterns and improving the forecasting of volcanic hazards. In an era where volcanic activity poses increasing risks to populations and infrastructure globally, leveraging multidisciplinary approaches is essential to enhance our ability to forecast eruptions and to assess volcanic hazards. By incorporating data from diverse sources—ranging from satellite platforms to ground-based sensors—researchers can build comprehensive models that better capture the complexity of volcanic systems. The session aims to highlight advances that are redefining how we detect, interpret, and respond to volcanic activity. Emphasis is placed on cross-disciplinary methods that couple remote sensing with machine learning, probabilistic frameworks, and impact assessment tools. We particularly encourage submissions that demonstrate advancement of knowledge in volcanology, near-real-time applications, scenario-based forecasting, and integration of diverse datastreams from ground-based and orbital platforms. By fostering collaboration across geophysics, computer science, and risk management, we seek to build a next-generation framework for volcanic hazard anticipation, response, and long-term resilience in the face of increasingly complex global challenges.