Zero Networks - Experts & Thought Leaders

Zero Networks, a provider of Zero Trust security solutions, announces results from its participation in the Anthropic Cyber Verification Program showing that a frontier AI model was unable to move beyond an intentionally exposed machine after gaining administrator access and extracting credentials. In the evaluation, Zero Networks’ microsegmentation prevented the model from breaching protected systems and identities, containing the attack to the initial point of compromise. The result addresses a growing security question as frontier AI models become increasingly capable of autonomous offensive activity: what happens when an AI attacker gets inside a network? Recent research from OpenAI showed AI agents discovering vulnerabilities, obtaining elevated privileges and ultimately contributing to the compromise of Hugging Face infrastructure. OpenAI researchers have warned that threat actors could eventually deploy similar autonomous agents against enterprises. Heterogeneous network environment Zero Networks was accepted into Anthropic’s Cyber Verification Program to conduct deeper security testing using Anthropic’s dual-use AI research capabilities. The evaluation put a frontier AI model against a realistic, heterogeneous network environment and tested whether it could use compromised credentials and offensive techniques to move laterally through the environment. The model successfully compromised the one machine deliberately left outside Zero Networks’ segmentation controls, gained administrator access and extracted credential material. But when it attempted to use those capabilities to reach protected assets, segmentation prevented further attacker movement. Across the evaluation, the model attempted 18 distinct attack paths using 23 offensive tools and touching 22 MITRE ATT&CK techniques, but did not move beyond the initially compromised machine. More consequential question In some runs, the model recognized that Zero Networks’ segmentation prevented access to additional systems and stopped pursuing further attack paths. “AI is going to make attackers faster, more persistent and increasingly autonomous. Security teams aren’t going to win by trying to out-click or out-think machines operating at machine speed,” said Benny Lakunishok, CEO and co-founder of Zero Networks. “The answer is to build environments where compromising one machine doesn’t give the attacker somewhere else to go. That’s what this test demonstrated: the attacker got in, had credentials and had every incentive to keep moving. The architecture stopped it.” Frontier models are increasingly capable of discovering vulnerabilities, obtaining credentials, reasoning through multi-step attack paths and operating with less human intervention. That makes the traditional question of whether an attacker can compromise an endpoint only part of the problem. The more consequential question: what that attacker can reach after the initial compromise? Zero Networks conducted its Anthropic evaluation to test the defensive side of that equation: if a highly capable AI attacker gets a foothold and acquires credentials, can architecture prevent it from turning that foothold into lateral movement? The evaluation showed that, in the tested environment, the answer was yes. Frontier AI model “What OpenAI disclosed should worry people well beyond the security team: autonomous agents finding each other, colluding, and escalating access without a human in the loop,” Dr. Chase Cunningham, a cybersecurity expert specializing in zero trust architecture. “The infrastructure the world actually runs on was never built to assume the attacker on the other end doesn’t get tired. Segmentation is one of the few controls that changes that math regardless of how the attacker thinks or how fast it moves, and this test is a concrete demonstration of why that matters.” Zero Networks built a large lab environment designed to resemble a production network, including Windows, Linux and macOS systems, Kubernetes and heterogeneous workloads. The company then applied Zero Networks microsegmentation to protect the assets and identities within the environment. One machine was intentionally left unsegmented to serve as the control and initial point of compromise. A frontier AI model was deployed as an autonomous attacker with administrator access and no restrictions on how it could attempt to attack the environment. The evaluation tested the model's ability to compromise systems, obtain credentials and pursue lateral movement using offensive security tools and techniques. Extracted credential material Claude attempted 18 distinct attack paths using 23 offensive tools and 22 MITRE ATT&CK techniques, including Kerberoasting, Active Directory credential extraction and a Pass-the-Hash attempt against the primary target. MITRE ATT&CK is the cybersecurity industry’s standard framework for cataloguing the tactics and techniques used by real-world attackers, so using 22 different techniques shows that Claude was not simply trying one exploit repeatedly, it was moving through a broad range of recognized attack behaviours, much like a human red team or sophisticated adversary would. Claude successfully compromised ESTI, gained administrator access, extracted credential material and planted the proof file. It was unable to use those capabilities to move into systems protected by Zero Networks segmentation.

Zero Networks, the provider of Zero Trust security solutions, announces a major expansion of its integration with Palo Alto Networks, extending the companies’ joint solution from automated microsegmentation and firewall policy orchestration to zero-touch containment and AI security. The original integration combined Zero Networks’ automated asset discovery, dynamic tagging, and identity-driven microsegmentation with Palo Alto Networks Next-Generation Firewalls. The expanded integration enables customers to use Zero Networks to identify and contain threats at the individual asset level, redirect selected traffic to Palo Alto Networks security services for deeper inspection, and manage enforcement through Strata Cloud Manager. The integration also brings together Zero Networks AI Segmentation and Prisma AIRS to govern AI agents and inspect AI traffic. Continuously mapping communication Rather than deploying separate controls for the perimeter, internal network, and AI environment, customers can apply consistent protection across on-premises, cloud, OT, Kubernetes and hybrid infrastructure. Zero Networks enables zero-touch containment by continuously mapping communication between assets and automatically creating least-privilege policies based on observed business activity. Through the expanded integration, selected traffic from protected workloads can be redirected to Palo Alto Networks firewalls for deeper Layer 7 inspection, enabling the platforms to block suspicious activity and contain affected assets before threats spread. Traffic redirection is available today for Linux environments, with Windows support planned, extending deep inspection and automated containment to Windows systems across the enterprise attack surface. Identity-based policies Zero Networks-managed assets and policies can now be surfaced through Strata Cloud Manager, giving security teams a consolidated view of discovered assets, segmentation policies, and the Palo Alto Networks controls inspecting their traffic. Dynamic asset tagging keeps policies current as workloads move, IP addresses change, and new assets appear, reducing the need to manually maintain static firewall rules. The expanded integration brings Zero Networks AI Segmentation together with Palo Alto Networks Prisma AIRS to provide security controls for AI agents and their traffic. Zero Networks identifies and governs AI agents, applying identity-based policies to restrict each agent to approved systems and destinations, while blocking unsanctioned AI services. When deeper inspection is needed, selected AI traffic can be redirected to Prisma AIRS for AI-aware threat prevention, combining control over where AI agents can connect with inspection of their prompts, responses, and data flows. Identity-based microsegmentation “Zero Networks controls which connections are allowed, Palo Alto Networks provides deep inspection,” said Benny Lakunishok, Co-founder and CEO of Zero Networks. “Together, we can contain malicious activity before it can move across the environment.” The joint solution enables organizations to: Contain compromised assets automatically with zero-touch, identity-based microsegmentation. Redirect selected traffic for Layer 7 inspection through Palo Alto Networks firewalls. Manage assets and policies through Strata Cloud Manager for unified visibility and enforcement. Govern AI agents and inspect AI traffic using Zero Networks AI Segmentation and Palo Alto Networks Prisma AIRS. Synchronize asset context with Palo Alto Networks Dynamic Address Groups. Dynamic asset tagging The integration uses existing Zero Networks and Palo Alto Networks enforcement capabilities, allowing customers to expand protection without redesigning the network or manually writing thousands of segmentation rules. The expanded integration between Zero Networks and Palo Alto Networks is available now, including automated microsegmentation, dynamic asset tagging, policy synchronization, and Linux traffic redirection.

Zero Networks, the foremost provider of Zero Trust security solutions, announced Least Agency Enforcement, a new capability that applies the Open Worldwide Application Security Project’s (OWASP) principle of Least Agency to help organizations safely deploy AI agents. Lateral Movement Exposure Report Zero Networks’ 2026 Lateral Movement Exposure Report research reveals that nearly 80% of enterprises have already deployed internal AI agents, yet two-thirds lack governance policies for them — creating rapidly expanding, unmanaged attack surfaces. As organizations grant AI agents the ability to access enterprise systems, invoke privileged tools, and make decisions with minimal human oversight, they must also define and enforce appropriate levels of autonomy. The OWASP Agentic Applications Top 10 Project’s Least Agency principle recommends constraining an agent’s autonomy, tool access, and decision-making authority to reduce the impact of prompt injection, privilege abuse, and compromised AI agents. Least Agency Enforcement Using identity-based microsegmentation, policy automation, and just-in-time MFA for privileged access, Zero Networks’ Least Agency Enforcement ensures that AI agents communicate only with explicitly authorized systems, only access approved resources, and require human approval before performing sensitive administrative actions. Unlike governance frameworks that document policy, Zero Networks enforces Least Agency across cloud, on-premises, Kubernetes, IoT/OT, and hybrid environments. “Least privilege works because it is simple: give people access to what they need, nothing more. We're doing the same thing for AI agents, except now it must be automatic, because nobody has time to babysit a thousand agents by hand,” said Benny Lakunishok, CEO and Co-founder of Zero Networks. “If an agent gets fooled or misused, it should hit a wall almost immediately, not wander around the network looking for something valuable. That's the bet we're making: less freedom for the agent now, which beats explaining a breach later.” With Zero Networks’ Least Agency Enforcement, organizations can: Limit AI agents to only the systems and services required for their assigned task Prevent lateral movement between applications, infrastructure, and administrative systems Require Just-in-Time MFA before AI agents access privileged ports or sensitive infrastructure Automatically generate and enforce least-privilege communication policies without manual rule creation Contain compromised, manipulated, or over-permissioned AI agents before they can impact critical business systems Least Agency Enforcement builds upon Zero Networks’ AI Security platform, including AI Agent Control, AI Segmentation, AI SaaS Control, and protection for enterprise LLM deployments. Together, these capabilities provide organizations with an enforcement layer that allows AI innovation to proceed without sacrificing security or cyber resilience. Availability Least Agency Enforcement is available now.  Zero Networks at Black Hat USA Zero Networks will showcase Least Agency Enforcement and its AI security platform at Black Hat USA 2026 in Booth #1852.