The integration of Terrestrial Networks (TN) and Non-Terrestrial Networks (NTN), driven by advancements in 3GPP Release 17 and beyond, is critical to achieving global connectivity in 6G communication systems. However, the dynamic and heterogeneous nature of TN-NTN systems introduces significant challenges, such as mobility-induced Doppler effects, latency constraints, and varying spectrum conditions. This chapter explores AI-enabled dynamic resource scheduling as a solution to these challenges, leveraging Deep Reinforcement Learning (DRL) and federated AI approaches to optimize resource allocation across multi-layered TN-NTN networks. We examine novel hierarchical AI frameworks for adaptive spectrum management and dynamic scheduling that address time-varying user demands and orbital dynamics of satellites. The chapter further discusses AI-driven techniques for mitigating interference, improving latency, and enhancing spectral efficiency in Low Earth Orbit (LEO) satellites, High Altitude Platforms (HAPs), and Unmanned Aerial Vehicles (UAVs). Key case studies demonstrate the efficacy of AI in achieving near-optimal throughput and fairness, while highlighting its computational advantages over traditional optimization methods. By incorporating AI into resource scheduling, this work paves the way for resilient and efficient TN-NTN systems, essential for next-generation wireless communication scenarios such as smart transportation, environmental monitoring, and ubiquitous broadband services.