Aegis of Tomorrow: An AI & Security Summit
Join a select group of AI practitioners and security leaders for an exclusive, closed-door discussion at the intersection of AI and software security.
We'll be joined by:
Daniel Miessler, AI & Security Researcher Extraordinaire
Jason Haddix, Hacker, CISO, CEO, & AI Specialist
Ken Huang, CISSP, AI Book Author, Speaker, & GenAI Security Instructor
Ken Johnson, CTO, DryRun Security
Seth Law, CEO, Redpoint Security
... and more!
Topics:
The Rise of AI Agents and Security Considerations
Speaker: Ken Huang
Ken Huang will discuss the emergence of AI agents and their potential impact on security. He'll covers key trends driving the rise of AI agents, including new scaling laws, the end of SaaS, physical AI, code agents, and agentic AI frameworks. Ken will also identify top agentic AI threats and provide a framework for threat modeling. Finally, he'll explore the importance of zero trust for agentic AI systems and offers actionable steps for securing AI agents, including identity management, data security, secure communication, red teaming, data reasoning without exposure, and blockchain-based trust. We will also cover an open source framework/SDK for Agent frameworks.
AI Attack Surface: Methodologies for Comprehensive Offensive Security Testing
Speaker: Jason Haddix
Jason will focus on a structured methodology for testing AI systems beyond just the model, including orchestration layers, integrations, and privilege escalation paths. He'll introduce detailed prompt injection taxonomy—breaking attacks down by intent, technique, evasion, and utility—with over 100 categorized examples from both academic research and the underground hacker scene. This approach gives security professionals a reproducible framework for uncovering real-world vulnerabilities in complex AI environments.
AI Capabilities for Attack and Defense (AICAD): A Framework for Prioritizing AI Spend on Cybersecurity
Speaker: Daniel Miessler
There is so much hype out there, and so many possible tools, that it’s hard to find clear signal on what to prioritize and actually build to improve your security program. This talk discusses an approach that takes current and future Attacker Capabilities as an anchor point. It looks at existing and desired attacker capabilities and asks the question of how our defenses will respond as they move through the AI tech tree in coming months and years. Includes supporting concepts as well as a number of attacker capabilities from the project.
Due to the limited space, please only apply for a seat if you are able to come to the event.
The hosts for the event are Redpoint Security, DryRun Security, Cannage Capital, and Hinge Health.