Learn directly from ai-security-in-action

This is a fantastic topic to dive in -- and the classes were fantastic! Well-explained theory and great hands-on practice via CTF made this course a one-of-a-kind. Would recommend!

Abhinav provides a great introduction to practical AI security. From introducing LLMs, to talking about AI Safety and Security, and threat modeling AI systems. The labs were interesting as they demonstrate direct and indirect prompt injections, jailbreaking, agentic security, scaling AI red teaming.

Abhinav is a good educator and the course material is very well done. As a novice in this field, it was easy to follow. The labs, especially, are challenging. After course completion, I like that everyone is still connected on Discord for discussions and sharing.

The course was incredibly insightful, providing a deep dive into the intersection of cybersecurity and AI. It helped me understand the inner workings of LLMs through a security lens, learning how to both attack and defend AI models. The cohort itself was a fantastic mix of professionals from both the AI and cybersecurity sectors. Crucially, this course focuses more on ‘Security in AI’ than simply ‘AI in security.’ The pace Abhinav maintains is, dare I say, PERFECT. He provides just enough time to truly grasp and contemplate the information being shared. We deconstructed LLMs to understand their underlying architecture, learning how to both exploit and fortify them. Abhinav has curated the ideal balance of theory and CTFs, allowing us to understand the nuances before seeing them in action through his well-structured labs. The ROI was universal. I’m heading back to the SOC with a significantly expanded threat model and the distinct advantage of having tackled these vectors early in my career. Cheers to Abhinav for putting together such a stellar course!

This course was very informative and helped me understand the security risks related to AI systems and LLM agents. The hands-on challenges were interesting and helped me learn concepts like prompt injection, agent misuse, data exposure, and tool exploitation in a practical way. Overall, the course improved my understanding of AI security and how vulnerabilities can appear in real-world AI applications.