Artificial intelligence and machine learning are rapidly transforming subsurface engineering by enabling more accurate, interpretable, and physics-consistent predictions across reservoir characterization, geomechanics, well log interpretation, and flow simulation. This session brings together recent advances in physics-guided machine learning, transformer architectures, agentic AI systems, digital rock workflows, and uncertainty-aware modeling approaches. The presentations will highlight innovative methods for permeability prediction, lithofacies classification, pressure transient analysis, borehole image interpretation, geological parameterization, and multi-phase numerical simulation. Emphasis will be placed on interpretable AI, integration with physics-based models, and practical applications for heterogeneous clastic and carbonate reservoirs. This session aims to bridge research and field implementation while demonstrating how AI-driven workflows are shaping the next generation of subsurface engineering.