Building on the concepts established in Part 1, Part 2 focuses on the practical application of AIdriven approaches and advanced analytical methods to strengthen Cybersecurity Situational Awareness (CSSA) within IoT environments. As IoT technologies continue to expand across sectors, this Part 2 examines the increasingly complex and dynamic security challenges arising from large-scale connectivity. The discussion extends and refines earlier methodologies, emphasizing the development of adaptive, intelligent, and scalable defense mechanisms that leverage AI and ML. A central theme is the investigation of AI-based threat and anomalydetection models designed for heterogeneous IoT networks. These models enable automated identification of abnormal activities and support early prediction of potential security weaknesses, allowing for proactive risk mitigation. The use of deep learning and neural network architectures for real-time analysis of high-volume IoT data streams is explored, demonstrating their advantages over conventional security assessment methods in terms of accuracy and responsiveness.