From geothermal resource development and carbon capture and storage to subsurface hydrogen storage and enhanced oil and gas recovery, conventional physics-based simulation workflows remain computationally prohibitive. This session invites contributions that advance AI and machine learning methods applied to the full spectrum of subsurface geophysical and reservoir engineering challenges. Topics of interest include, but are not limited to deep generative models, foundation models, agentic AI, and reinforcement learning — applied to subsurface characterization from sparse or noisy data, seismic wave modeling and inversion, surrogate modeling of reservoir dynamics, uncertainty quantification, and closed-loop reservoir management of geo-energy system design. We particularly welcome work that integrates domain physics with data-driven approaches to improve interpretability, generalizability, and operational reliability. By convening ML researchers, reservoir engineers, and energy systems practitioners, this session aims to accelerate knowledge transfer between communities and chart a roadmap for AI-driven geo-energy science.