This course examines the evolving landscape of security threats targeting Artificial Intelligence (AI) and Large Language Models (LLMs), with a focus on cutting-edge attack vectors. Students will analyze how these methods exploit system vulnerabilities and gain practical skills to assess and defend against such risks. Emphasis is placed on the application of the MITRE ATLAS framework to evaluate real-world threats and inform mitigation strategies. Through hands-on activities and case studies, students will design adaptive security models that support continuous monitoring, threat detection, and operational resilience in AI-driven environments.