Building a Security Blueprint for the Ungoverned Agent Problem

Every week, employees across your organization are creating AI agents that are automating workflows, connecting to systems, and acting on behalf of the business. Most security programs have no record they exist.

These agents don’t go through onboarding. They don’t appear in your identity lifecycle. They inherit credentials from the people who built them and start operating quietly, under valid access, outside any governance model. The risk isn’t theoretical. It’s already running.

For CISOs and security practitioners responsible for AI strategy, risk, and governance, the question is no longer whether to adopt AI, it’s whether your security has kept pace with how fast it’s moving.

Join Mark Clancy (CISO, T-Mobile), Jason Lish (CISO, Cisco), and Ryan Knisley (former CISO, Walt Disney and Costco), for a candid conversation on building the security program that makes confident AI adoption possible.

What you’ll take away:

• A clear picture of where AI agent risk lives in the enterprise today, and why traditional controls don’t see it
• The governance questions your board and regulators will ask next, and how to get ahead of them
• How security leaders at large, complex enterprises are approaching agent governance without putting the breaks on innovation
• What a defensible AI security looks like in 2026, and what to prioritize next