Artificial intelligence (AI) is emerging as a transformative tool for next-generation wireless communication systems, enabling enhanced security, efficiency, and adaptability. This project investigates two complementary research directions: covert communications and intelligent resource management. In the first, we apply machine learning (ML) methods (including reinforcement learning and generative models) to design adaptive signaling and spectrum-shaping strategies that minimize detectability while maintaining reliable performance in adversarial environments. In the second, we explore AI-driven approaches for dynamic spectrum allocation, power control, and interference management, targeting more efficient use of scarce wireless resources in dense and heterogeneous networks. By integrating learning-based optimization with communication theory, the project aims to establish a unified framework where AI enhances both the stealth and efficiency of wireless systems. Anticipated outcomes include novel algorithms, performance benchmarks, and practical insights that advance the state of the art in secure, intelligent, and resource-efficient wireless communications, with potential applications spanning critical infrastructure, defense, and civilian networks.