AI vs AI: A Live Demonstration of Defensive Agentic Countermeasures to Adversarial AI
Join us for breakfast and a closed door exclusive demonstration of agentic countermeasures
Large Language Models now power both sides of the email threat landscape. Attackers use autonomous agents to generate adaptive phishing and impersonation campaigns. Defenders deploy LLM-based systems to analyze intent, detect manipulation, and respond at machine speed.
In this session, we will move beyond theory.
We will conduct a live, controlled evaluation in which adversarial and defensive LLM agents operate against each other in real time across simulated enterprise email scenarios.
We will examine:
How quickly adversarial agents adapt to defensive controls
Where defensive LLMs meaningfully detect or fail
The impact of prompt manipulation and model drift
Whether “AI vs. AI” structurally favors offense or defense
This session is designed for CISOs seeking empirical insight into how autonomous systems reshape the email attack surface.