Aviatrix Launches AI Agent Containment Security Platform

Aviatrix has launched a platform to contain AI agents and enforce security controls across cloud workloads without requiring code changes. The move responds to rising supply chain attacks and threats like prompt injection and model poisoning that can compromise AI agents operating across enterprise systems. The platform extends Aviatrix's Cloud Native Security Fabric with two products: Zero Trust for AI Workloads, now generally available, and AgentGuard, in early access. CEO Doug Merritt emphasized reducing blast radius as the key security metric, comparing the approach to dividing a castle into locked rooms so a breach in one area cannot spread throughout the network.