With the introduction of AI, and AI agents, AI- and AI agent hacking became a thing, almost instantly.

In this talk we'll take a look at how AI agents work, and how they empower generative AI. Then we will showcase hacking techniques to reveal their vulnerabilities and demonstrate how they can be leveraged with catastrophic results.

Drawing from real-world experience, we’ll explore methods used to exploit these systems, uncovering the risks they face.

Finally, we’ll wrap up with practical strategies to defend against these kinds of attacks.