Xage extends zero trust to autonomous AI agents across cloud, SaaS and edge

Zero-trust cybersecurity company Xage Security Inc. today unveiled new capabilities in its platform designed to give enterprises deterministic visibility into autonomous artificial intelligence agents and block them from taking unauthorized actions when compromised.

The release introduces two components, Xage Agent Sentry and Xage Resource Gateway, that the company says together can wrap an AI agent and the resources it touches with controls operating at the network interaction, local event and operating system call levels. The pitch is that prompt-level guardrails are not sufficient once agents are connected to live databases, software-as-a-service tools and operational technology environments.

Xage Agent Sentry wraps an agent wherever it runs and monitors everything moving in and out of it. Xage Resource Gateway sits in front of critical resources and governs how AI systems interact with them. Each agent is assigned a digital identity at onboarding so security teams can attach role, resource and time-bound policies and the platform also flags unmanaged shadow AI agents so they can be onboarded or removed.

The capabilities build on the company’s previously announced Zero Trust for AI work covering Model Context Protocol and agent-to-agent communication and on its October integration with the Nvidia Corp. BlueField data processing unit aimed at securing AI factories. Xage said the new release extends coverage to SaaS applications, cloud services, on-premises systems and the edge.

In a demonstration accompanying the launch, Xage showed an OpenClaw agent being hacked and then blocked by its Zero Trust for AI platform from exfiltrating data or damaging organizational resources. The company said the architecture supports closed-loop, long-running agents that operate without constant human approval, with the option to keep a human in the loop.

“AI is ready to move beyond the sandbox, but enterprises cannot safely deploy it in production unless they know exactly what agents are doing and can control the actions they take,” said Chief Executive Duncan Greatwood. A June 2025 projection from Gartner Inc. suggested that 40% of AI projects would be canceled by 2027 due to inadequate risk controls.

The company also pointed to anomaly detection features built on agent activity logs, including behavioral baselining to catch deviations such as an agent that normally only reads suddenly issuing write commands, plus integration with security information and event management and security operations center tooling

James O’Keefe, vice president and strategist, solutions and technology group, and chief technology officer at Science Applications International Corp., said that federal and defense agencies “need unified visibility, unimpeachable control, and continuous oversight of agent activity across classified and unclassified environments” as AI agents move into mission-critical operations.

The new capabilities are available now and extend Xage’s Fabric Platform, which the company markets as deployable in a day across cloud, data center and edge environments.

Image: Xage