At the 2025 RSA Conference, Justin Foster joins Zeus Kerravala to break down where AI in cybersecurity is actually delivering value and where it’s falling short. As security teams deal with growing complexity, many are finding that today’s AI tools create as much friction as they solve. This conversation explores how a shift toward agentic, skills-based AI can help teams move faster, reduce noise, and focus on what really matters.

Claude Mythos is an AI model capable of finding and chaining zero-day vulnerabilities at scale. That changes how attacks happen, especially in environments where you can’t patch fast enough. The Forescout 4D Platform with VistaroAI helps organizations respond with real-time visibility and dynamic control across all connected devices.

See how ServiceNow and Tanium AI proactively detect, diagnosis, and resolve VPN issues in real time.

AI adoption is accelerating across enterprises, often faster than security teams can respond. Employees are already using AI tools, copilots, and agents across SaaS apps, browsers, and workflows. That creates new risk around shadow AI, sensitive data exposure, runtime threats, and autonomous actions that traditional controls were never built to handle. In this video, we break down the four AI security challenges CISOs are facing right now.

Autonomous AI agents are no longer experimental—they’re writing code, executing commands, and making decisions in real time. But as AI coding agents become more powerful, they’re also introducing a new and often invisible attack surface. In this video, we break down: AI agents can install packages, run scripts, and modify systems instantly—often without traditional visibility. That means security teams need to rethink how they monitor and protect their environments.