GeoResourceCloud is an agentic and data-driven platform for multiscale and multiphysics modeling of future subsurface energy systems. The platform integrates flow-geomechanical upscaling, coupled THMC simulation, uncertainty quantification, optimization, and AI agents to support decision-making under geological uncertainty. Using geological CO₂ storage as an example, GeoResourceCloud can couple reservoir–geomechanics models with reinforcement learning to optimize injection strategies while controlling pressure buildup, assessing caprock integrity, and reducing leakage or geomechanical risks. AI agents further automate workflow execution, model interpretation, tool calling, and real-time decision support through natural-language interaction. By combining physics-based simulation, machine learning, and reliable AI agents, GeoResourceCloud aims to improve the safety, efficiency, and sustainability of subsurface energy development.

Multiscale;Multi-Physics;Data-Driven;Agentic;GeoSystem