TLDR: Xage Security has launched a groundbreaking Zero Trust Platform designed to secure artificial intelligence (AI) systems by enforcing policies at the fundamental protocol layer. This innovative approach aims to eliminate “jailbreak” risks and provide robust, identity-first controls for AI agents and workflows, addressing the critical security gaps in current AI deployments.
Palo Alto, CA – September 9, 2025 – Xage Security, a recognized leader in Zero Trust access and protection, today announced the release of its pioneering Zero Trust for AI platform. This new offering is specifically engineered to secure AI environments by implementing policy enforcement at the deepest protocol layers, a significant departure from conventional, often vulnerable, AI security measures.
The rapid integration of AI into critical business operations, handling sensitive data, running complex workflows, and making crucial decisions, has exposed inherent weaknesses in existing security paradigms. Traditional safeguards such as prompt filters, Large Language Model (LLM) firewalls, and siloed deployments are proving to be reactive, easily circumvented, and expensive to maintain. These methods operate at the prompt or output level, making them susceptible to bypass through creative inputs, adversarial prompts, or chained queries.
Xage Security’s new platform tackles these challenges head-on by enforcing identity and policy at the network data protocol layer. Duncan Greatwood, CEO at Xage Security, emphasized this distinction, stating, “Jailbreak-proof means Xage enforces identity and policy at the network data protocol layer, not just at the AI prompt or AI output layer… Even if AI prompts are manipulated, unauthorized access is blocked because policy enforcement happens below the AI layer, where it can’t be bypassed.” This strategic shift redefines AI security, moving the focus from merely filtering what models generate to rigorously controlling what AI agents and applications are authorized to do.
A cornerstone of the platform is the “Identity reimagined for AI agents” concept. Recognizing that AI agents operate at machine speed, spinning up connections to APIs, databases, and external tools, Xage extends its proven Zero Trust Fabric to encompass these non-human identities. Each AI agent is assigned cryptographic credentials, scoped entitlements, and subjected to automated rotation, ensuring that every interaction is governed by principles of least privilege and continuous behavioral monitoring. This prevents blind spots and mitigates risks associated with ungoverned AI behavior.
The Xage Fabric Platform provides comprehensive, full-stack security, safeguarding every layer of the AI and data center infrastructure. This includes locking down access to physical hardware like servers, GPUs, and edge devices, enforcing least-privilege controls on training, inference, and orchestration layers, and governing how LLMs, agents, and users interact with sensitive data.
Mark Gudiksen, Managing Partner at Piva Capital, underscored the urgency and importance of such robust solutions: “AI is being embraced at a pace that rivals the early days of internet adoption—only faster, deeper, and across every industry. But with that momentum comes risk. We’ve already seen examples in the news of what happens when AI systems operate without the right safeguards. The long-term success of AI depends not just on innovation, but on the rigorous controls needed to govern it. Control isn’t optional—it’s the foundation for safe, scalable AI and the enabler for universal AI adoption.”
Also Read:
- Vanta Unveils Advanced AI Capabilities for Proactive Security and Compliance
- OneTrust Unveils AI-Ready Governance Innovations Amid Surging AI Risk Management Demands
The platform is designed to be lightweight and deployable across any environment, from cloud to data center to the remote operational edge, allowing enterprises to rapidly implement Zero Trust security and access controls. By anchoring AI security in the same principles that protect critical infrastructure, Xage aims to instill confidence in enterprises and Managed Security Service Providers (MSSPs) to safely scale their AI initiatives.
