Straiker announced two products at RSAC 2026 targeting security gaps in enterprise AI agent deployments. Discover AI provides agent inventory and risk detection, scanning for AI agents already running in enterprise environments, detecting MCP vulnerabilities from a database of 12,000+ entries, and flagging risky agent behavior. Defend AI has been expanded with training on millions of real-world agent traces, achieving sub-300ms latency and 98%+ detection accuracy, with new integrations for Amazon Bedrock AgentCore, Azure AI Foundry, and Microsoft Copilot Studio. Both products address the growing risk of AI agents operating with broad access and minimal security oversight across enterprise systems.
Sort: