AI Agent Security Summit: Assembling to shape the Future of Autonomous Defense

Enterprises are rapidly deploying AI agents that don’t just process data, they act. These agents connect to email, calendars, development environments, and production systems. The attack surface hasn’t simply grown, it has multiplied.

The security community must adapt quickly. The AI Agent Security Summit is where the best and brightest assemble to share research, compare real-world incidents, and shape the future of securing autonomous systems. This isn’t just another conference, it’s a gathering of the leaders defining how we defend against the next wave of threats.

Real-World Exploits, Not Hypotheticals

Ben Nassi’s research on Promptware shows how attacks have already evolved into malware with real financial consequences, from supply chain compromises to persistent “ZombAIs” operating as part of attacker infrastructure.

Johann Rehberger’s Month of AI Bugs complements this by demonstrating how easily enterprise agents across Microsoft, Google, OpenAI, and Salesforce can be hijacked. His findings reveal just how fragile current authentication and authorization controls really are.

These sessions make one thing clear: AI agent threats are active, not theoretical, and only by learning from each other can we get ahead of them.

Practical Defense Strategies for Autonomous Systems

The summit goes beyond identifying the problem. Together, we’ll explore and pressure-test actionable defenses, including:

• Insider threat frameworks for treating AI agents as “employees” who must be governed and monitored.
• Autonomous patching systems that detect, generate, and validate fixes at machine speed.

• AI Red Teaming methodologies that turn adversarial techniques into defensive assets. David Campbell will demonstrate how adversarial red teaming fortifies AI applications across every layer, while Ken Huang's AIVSS framework incorporates red teaming for vulnerability validation
• Governance models that balance innovation with safety.
• Behavioral monitoring frameworks that predict and prevent emergent risks before they become incidents.

The AI Agent Security Summit convenes the brightest thinkers and doers in the field to tackle today’s hardest challenge: securing autonomous agents at enterprise scale. By pooling expertise across research and practice, the summit builds a defense roadmap that simply cannot be written in isolation.

Why You Should Join

The parallels to early web application security are unmistakable: over-trusted inputs, weak boundaries, and assumptions that systems behave predictably. The difference now is scale and impact. When an application fails, you lose data. When an AI agent is compromised, you lose control of a system that can act across your enterprise.

The AI Agent Security Summit is more than a set of sessions, it’s a community of researchers, practitioners, and enterprise leaders shaping the next era of security together. By joining, you’re part of a collective effort to build the guardrails that will define how enterprises adopt and secure autonomous agents for years to come.

Register Today

The AI Agent Security Summit is where the security community comes together to shape the future of defending autonomous systems. Seats are limited, secure yours in San Francisco on October 8th.