Xeris Revolutionizes AI Agentic Cybersecurity With Patent Pending Super AI Agent Technology

Super AI Agents bring deterministic security to non deterministic, self modifying AI environments

Xeris Revolutionizes AI Agentic Cybersecurity With Patent Pending Super AI Agent Technology

Tel Aviv, Israel, Xeris today announced a major breakthrough in AI Agentic cybersecurity with its Patent Pending Super AI Agent technology, purpose built to secure the rapidly expanding world of autonomous AI Agents, MCP based workflows, and self evolving AI systems operating inside the enterprise.
As organizations accelerate adoption of AI Agents that can reason, act, generate code, and dynamically extend their own capabilities during runtime, existing security architectures are failing to keep pace. Security solutions designed for static applications, APIs, and human driven workflows lack the visibility and control required to protect these highly dynamic, non deterministic environments.

While Agentic abilities are incredibly productive for enterprises, they also create a greenfield for a new class of attacks, more dangerous, and near impossible to detect with traditional tools.”

— Shlomo Touboul

“AI Agents are no longer experimental, they are becoming core enterprise actors,” said Shlomo Touboul, Founder of Xeris. “While these abilities are incredibly productive for enterprises, they also create a greenfield for a new class of attacks, more sophisticated, more dangerous, and near impossible to detect with traditional security tools.”

Xeris was founded from day one to address this exact challenge.

The company introduces a new security control layer for the Agentic era, the Super AI Agent. This autonomous supervisory AI continuously observes, understands, and governs other AI Agents in real time. Rather than relying on static rules, predefined policies, or post event detection, Xeris Super AI Agents adapt at the same speed as the AI systems they protect.

“When AI can change itself during runtime, legacy security becomes blind,” added Touboul. “Xeris Super AI Agents were designed to supervise, control, and secure AI Agents at scale, without slowing innovation.”

Also Read: AiThority Interview with Zohaib Ahmed, co-founder and CEO at Resemble AI

A New Category of AI Native Security
Xeris Super AI Agents operate above and across the entire enterprise AI ecosystem, including AI Agents, MCP servers, tools, data sources, and identity layers. They deliver continuous runtime visibility into AI behavior, and uniquely, the Super AI Agent itself is an AI Agent that can evolve during runtime. As the protected AI Agents extend their capabilities, create new tools, or modify their behavior, the Super AI Agent dynamically extends its own understanding and control functions to match those changes, ensuring security coverage remains complete even as the AI environment reshapes itself in real time.
This Patent Pending approach enables deterministic control in an inherently non deterministic environment, a critical requirement as AI Agents increasingly make autonomous decisions that directly impact data integrity, business operations, and organizational risk.

Built for the Agentic Era
Unlike legacy security platforms retrofitted from API gateways, IAM, or traditional XDR technologies, Xeris was designed from the ground up for Agentic environments. Its architecture is optimized for hyper growth AI adoption, dynamic system evolution, and enterprise scale governance of autonomous AI behavior.

Key Capabilities Include
• Continuous runtime visibility across all AI Agents and MCP interactions
• Autonomous supervision and enforcement using Super AI Agents
• Enterprise wide policy control across dynamic, self modifying AI systems
• Elimination of Shadow AI and unauthorized AI driven actions
• Scalable, low latency architecture built for large scale AI adoption

Also Read: The Death of the Questionnaire: Automating RFP Responses with GenAI

[To share your insights with us, please write to psen@itechseries.com]