Aviatrix launches AI agent containment platform to limit blast radius in cloud workloads

Aviatrix Introduces AI Agent Containment Platform for Cloud Workloads

Aviatrix has launched a new AI agent containment platform designed to enforce security controls and isolate AI agents across cloud workloads without requiring changes to existing AI agents or code 1

1

. The platform extends the company's Cloud Native Security Fabric with two new products: Zero Trust for AI Workloads, now generally available, and Aviatrix AgentGuard, currently in early access . This launch responds to an increasing number of supply chain attacks that can directly affect the operational security of agents, dependencies, and code logic itself.

Addressing Emerging Threats Like Prompt Injection and Model Poisoning

AI agents introduce unique security challenges because they don't need to be "broken into" in the traditional sense to become dangerous. An agent can be manipulated through prompt injection, where malicious instructions are hidden in content the agent reads, or through model poisoning, where the data or tools it depends on are corrupted 1

1

. If that agent has broad access to applications, files, credentials or external services, a successful compromise could allow it to move data, call tools or communicate with systems far beyond its intended role. CEO Doug Merritt argues that enterprises need to do more than simply detect and react to anomalies—they must contain AI workloads and isolate them from other systems to prevent compromised agents from breaking out of their assigned roles 1

1

.

Securing AI Agents by Limiting the Blast Radius

"My argument for the containment era is the most important metric is blast radius," Merritt told SiliconANGLE 1

1

. Rather than focusing solely on detection and remediation, the platform aims to shrink the blast radius associated with compromised AI agents 2

2

. For the vast majority of enterprises, there is no architectural constraint on where a compromised workload can go, meaning the blast radius can include all enterprise applications and even the entire network 2

2

. Merritt described the approach as creating "this beautiful honeycomb of things that communicate," where each workload has carefully defined communication pathways and "when something goes wrong in one cell, it doesn't affect the other cells" 1

1

.

Zero Trust for AI Workloads and Communication Governance

With Zero Trust for AI Workloads, information technology teams can secure AI agents, large language model proxies and agentic frameworks without requiring application or infrastructure changes 1

1

. The product allows teams to set policies that allow or deny access to external AI services, block shadow AI with allowlists, and apply network-layer enforcement across workloads and regions. It addresses one of the central problems of agentic AI security: workloads often need to communicate to do their jobs, but they should not be able to communicate with everything 1

1

. For AI agents, that line can be difficult to draw because agents may behave like users in one moment and like applications or services in the next. As Merritt put it, "An agent is weird because it's kind of half-human, half-workload" 1

1

.

How Aviatrix AgentGuard Works to Contain AI Agents

Aviatrix AgentGuard provides full containment by discovering every agent running across virtual machines, Kubernetes clusters and serverless functions, both authorized and shadow 1

1

. It maps LLMs, tools and the data each agent connects to, then builds an updating risk profile 1

1

. Using that risk profile, AgentGuard monitors activity and automatically blocks behavior that does not match the agent's baseline. Behaviors that match common exfiltration patterns, such as posting data to public code repositories or file-sharing services, are blocked by default 1

1

. AgentGuard extends the Cloud Native Security Fabric to enforce communication governance at the agent workload level, governing what each agent can reach and what can reach it 2

2

.

The Economics of Cyberattacks and the Containment Era

Merritt argued that the economics of cyberattacks are changing as AI models make sophisticated attack techniques easier to automate and scale. "We are democratizing capabilities for nefarious behavior at the same time that we are completely changing the economics of nefarious behavior," he said 1

1

. That makes containment more urgent. If attackers can move faster, more cheaply and with more automation, then security teams need a way to limit how far a compromised agent or workload can go before detection even happens. "I believe deeply, as one of the fathers of the detect and remediate era, that we're entering a containment era," Merritt told CRN 2

2

.

Opportunities for Channel Partners in AI Security

For channel partners, the platform opens the door for add-on diagnostic and assessment services, focusing on blast radius assessments and agent governance 2

2

. Many enterprise IT teams still need help mapping their environment and understanding their full attack surface, including where their perimeters and egress points are, how to design stronger containment, and how to run that effectively day to day. "There is a significant services revenue stream about to be unleashed for channel partners that understand the dynamics that AI is bringing, and what that means for vulnerability detection or remediation, and why containment is likely to be a higher growth area over the next 1-2-3 years within these organizations," Merritt said 2

2

. Santa Clara, California-based Aviatrix released a revamped partner program in 2025 with a goal of driving 100 percent of its transactions through the channel 2

2

.