From prompt to exploit: How LLMs are changing API attacks

The way applications are being attacked is changing faster than most security teams realize. As LLMs enable faster discovery and exploitation of weaknesses, APIs have become the easiest path for attackers to navigate complex, interconnected systems.

The problem is not just more attacks, but also the speed and efficiency at which they now happen. Traditional security approaches were not built for automated exploit chaining or machine-speed reconnaissance, creating gaps in visibility and control.

This session is for anyone looking to understand how attacker behavior is evolving in practice, and what that means for securing modern, API-driven environments. It breaks down where those gaps exist and how teams need to rethink trust, access, and risk in an AI-driven world.