Secure access service edge platforms are emerging as a potential answer to the growing mismatch between legacy security architectures and modern AI workloads. But closing that gap may require more than modern SASE can handle — instead calling for a true AI-native SASE platform.

The SASE market is growing at a compound annual growth rate of 26%, but the next wave of growth depends on platforms that can secure AI-native workflows, not just traditional traffic. Cato Networks Ltd. is positioning itself at the center of that shift with two major product launches, according to Nimmy Reichenberg (pictured), senior vice president of product marketing at Cato Networks Ltd. Those include Cato Neural Edge, a GPU-powered AI inspection infrastructure, and Cato AI Security, designed to protect enterprise AI use.

“When you combine those two announcements, really what you get is the first AI-native SASE platform,” Reichenberg told theCUBE. “Securing AI interaction is now native on the SASE platform. What that means is that we’re able to offer security for AI that you use, that your employees use, … for the AI-powered applications that companies are building themselves … and for agents.”

Reichenberg spoke with theCUBE’s Dave Vellante and Christophe Bertrand at the RSAC 2026 Conference, during an exclusive broadcast on theCUBE, SiliconANGLE Media’s livestreaming studio. They discussed how Cato’s AI-native SASE platform addresses the dual challenge of securing enterprise AI adoption and using AI to improve threat detection. (* Disclosure below.)

Building an AI-native SASE platform on GPU-powered infrastructure

The two announcements represent Cato’s bet that owning its own GPU-powered cloud, rather than relying on hyperscaler infrastructure, gives it a structural advantage. Cato Neural Edge embeds Nvidia Corp. GPUs across more than 85 points of presence worldwide, enabling inline AI inspection without routing traffic to external environments, Reichenberg explained.

“We’ve always believed that by owning our own cloud, we can provide a very resilient service to our customers, and we’re just bringing GPUs to our own cloud as opposed to using somebody else’s GPUs,” he said.

The AI security capability, built on technology from Cato’s acquisition of Aim Security Ltd. last year, was fully integrated into the platform in less than six months. That speed of integration underscores the advantage of a single-architecture approach, Reichenberg noted. In production, the platform uses what Cato calls its single pass cloud engine — all inspection happens in one place in one pass.

“The advantage to us is this whole seamless single pass inspection now extends to AI with the GPUs,” he said. “Everything’s faster, more streamlined and easier to manage.”

Agentic AI is amplifying the urgency to solve the problem. Traditional SASE principles such as zero trust carry over naturally to autonomous agents, but new requirements — such as inspecting what tools agents use and how Model Context Protocol servers are accessed — demand purpose-built controls, according to Reichenberg.

“A year ago, nobody asked us to secure MCP servers because they didn’t exist. Nobody asked us to secure agentic browsers because they didn’t exist,” he said. “Everything’s moving so fast, the only thing we can really do is [rely on the fact] that we’ve got the expertise, we’re committed to making the investments and I’m not betting against the fact that the hottest thing in AI security 12 or 18 months from now is something that none of us have heard of.”

Here’s the complete video interview, part of SiliconANGLE’s and theCUBE’s coverage of the RSAC 2026 Conference:

(* Disclosure: Cato Networks sponsored this segment of theCUBE. Neither Cato Networks nor other sponsors have editorial control over content on theCUBE or SiliconANGLE.)

Photo: SiliconANGLE