Conference Description
Key Takeaways
- Single-track, in-person summit addressing AI security, governance and risk management for enterprise environments
- Technical focus on agent identity, authentication protocols, probabilistic prompt injection and AI-powered attacks
- Designed for CISOs, CTOs, VPs of Security, Heads of AI and board-facing executives responsible for AI risk
- Practitioner-led sessions featuring incident reports, architecture decisions and red team findings
- Sponsors include Snyk, AWS, NVIDIA, Anthropic, Google, Microsoft, Okta and BlackRock
About the Event
The AI Security Summit SF 2026 is a single-track, in-person event bringing together practitioners who build AI systems and organizational leaders responsible for managing AI risk. The summit prioritizes technical depth and real-world trade-offs over vendor presentations, featuring keynotes, panels and fireside chats from security and AI professionals who have made critical decisions about AI risk at scale.
Content is peer-reviewed and practitioner-led, with speakers sharing incident reports, architecture decisions and defensive strategies drawn from production environments. A technical after-party includes live demos and lightning talks.
Technical Topics and Security Challenges
The summit addresses security concerns specific to AI systems, with particular attention to agentic AI architectures. Sessions cover agent identity and authentication, including discussion of evolving protocols beyond OAuth such as AAuth (agent authentication protocol). Probabilistic prompt injection attacks receive dedicated coverage, alongside broader examination of attacks operating at machine speed and corresponding AI-powered defensive measures.
Additional topics include trust in AI software delivery, securing data touchpoints in agentic systems, and the security implications of open source models. The programme connects these technical challenges to enterprise risk management, examining how organizations make investment decisions and set risk budgets for AI deployments.
Enterprise Governance and Risk Management
Beyond technical implementation, the summit addresses organizational challenges in AI governance. Sessions examine how enterprises determine which AI risks to defend against and which to accept, how to evaluate and select vendors, and how to communicate AI risk to boards. The event recognizes that much current AI security knowledge resides within companies rather than in published research, making practitioner knowledge sharing central to the programme.
Who Should Attend
The summit targets senior leaders who own AI risk at the organizational level. This includes CISOs, CTOs, VPs of Security and Heads of AI, as well as security architects, AI engineers and technical leads. The event is particularly relevant for those working in large enterprises, financial institutions and technology companies deploying AI at scale, where decisions about AI governance carry significant business and regulatory implications.