Session 1: Towards Building Safe and Secure Agentic AI

This session explores the security challenges and opportunities of agentic AI, focusing on how autonomous systems expand attack surfaces while enabling powerful defenses. Grounded in real world zero‑day results, we examine risks of code-generating agents, demonstrate real-world exploits and defenses, and share practical safeguards and governance strategies for secure deployment.

Speaker: Dawn Song, Professor, University of California, Berkeley

Session 2: The Paradox of Agentic AI: Scaling Security While Eroding Talent

This session explores the paradox of agentic AI: driving scale and access in security while eroding talent and introducing a new insider threat. As a father, former CTO, and CEO defending 600,000 organizations worldwide, Levy offers strategies to secure AI as critical infrastructure, align agentic systems with intent, and prepare teams for a future where intelligence is the new battlefield.

Speaker: Joe Levy, CEO, Sophos

Session 3: Agentic Role Mapping: A Field Guide for Secure AI Oversight

AI agents are entering environments fast, but who’s accountable when they act? This session introduces Agentic Role Mapping, a pressure testing a framework that helps ISSMs, ISSEs, and security teams define oversight, restructure responsibility, and build cognitive guardrails before AI outpaces policy.

Speaker: Allen Westley, Director, Sr. Principal Cyber Intelligence, L3Harris Technologies

Session 4: My AI Will Call Your AI: Securing Agent-to-Agent Communications

AI agents will soon handle tasks ranging from booking flights to negotiating on our behalf, often by communicating directly with other agents. This creates a new security challenge: securing agent-to-agent communications. We will show how these systems can be exploited and share practical IAM and SOC strategies to defend them.

Speakers: Kanna Sekar, Senior Customer Engineer, Security and AI, Google and Ravi Karthick Sankara Narayanan, Senior Security Consultant, Deloitte

This virtual seminar is sponsored by Wiz. Wiz transforms cloud security for customers with an operating model built to automate risk reduction and threat response, enabling security teams to operate at AI speed.