Lufsec AI Hacking is a training course focused on teaching ethical hacking, testing, and securing large language models and AI systems. The course consists of 5 chapters and 7 lessons covering topics from basic prompt manipulation to advanced adversarial machine learning techniques. The curriculum includes foundational concepts such as prompt hacking, jailbreaking, and adversarial ML, along with ethical considerations and responsible disclosure practices. Students learn through hands-on labs that demonstrate real-world attack scenarios and defense strategies. The course covers prompt injection and jailbreaking techniques, context and cognitive attacks, advanced prompt exploitation, and LLM application vulnerabilities. Each module includes practical labs such as prompt filter exploitation, inducing malicious behavior, crafting adversarial prompts, and red teaming vulnerable applications. The training aligns with modern security frameworks including OWASP for LLMs and Google's Secure AI Framework (SAIF). The course aims to help learners identify AI security weaknesses and develop skills to defend and build resilient AI systems. A free preview is available for select lessons before full enrollment.