Cato Networks Acquires Aim Security for AI Threat Protection

Cato Networks purchased a startup led by the Israel Defense Forces’ former head of research and development to embed artificial intelligence security directly into the network layer.

See Also: Securing the ‘Gateway to Identity’ With an Enterprise Browser

Tel Aviv, Israel-based secure access service edge provider will integrate Aim Security into its Single Pass Cloud Engine to provide seamless protection across internally developed, agentic and third-party AI systems, according to co-founder and CEO Shlomo Kramer. AI security is becoming a foundational new layer in network security on top of traffic and data/application protection, Kramer said.

“AI is not just another layer of security, it’s a whole new stack that is completely different than previous stacks,” Kramer told Information Security Media Group. “It’s listening to thousands of conversations in English and deciding what is okay for the enterprise and what is not. Both the capabilities and the rules of reaching that decision are completely new.”

Aim Security, also located in Tel Aviv, dates to 2022, employs 79 people and raised $28 million of outside funding, having most recently completed an $18 million Series A round led by Canaan Partners in June 2024. The firm has been led since inception by Matan Getz, who spent more than nine years in the IDF, leading to a role leading a 150-person R&D unit. Calcalist said Cato will pay between $350-to-$400 million for Aim (see: How AI Expands SASE’s Role in Policy, Operations Efficiency).

Why Cato Would Rather Buy Than Build

The AI security market is growing too quickly for Cato to spend years building a platform from scratch, Kramer said. Even if the company did build one, it would take too long to reach the technical maturity, research excellence and customer validation that Aim already possesses.

“This market is exploding at a pace that is unprecedented, and CISOs need a solution now,” Kramer said. “We calculated that the time to converge Aim is going to be a few months, and the time to reach the same level of depth and maturity and experience and research and all these this team has done in the last three years would take us years. It was an easy decision to go on the buy side.”

Cato Networks is uniquely positioned to deliver AI security since its SASE architecture already serves as a central hub for traffic and data flows, Kramer said. AI systems often work by communicating across the network – with LLM-powered applications or autonomous agents calling APIs – and Cato’s position in the middle of that network layer gives it unmatched visibility and influence, according to Kramer.

“It’s a huge opportunity for the expansion of the SASE envelope to include this new layer, because SASE is in the middle of all this conversation,” Kramer said. “We are the network security. All conversations happen over the network, and so we are in a fairly good place to deliver that new ability.”

Aim’s AI engine operates by analyzing traffic-level conversations, whether those are between users, applications, or autonomous AI agents, which Kramer said complements Cato’s architecture since it already inspects and routes enterprise traffic. The integration entails routing that traffic through Aim’s engine, capturing insights and returning actionable signals into Cato’s enforcement and policy systems.

“It mainly entails being able to integrate our AI security brain into our space, our ability to do a single pass on the traffic, pass it on to their engine, get back the results and act upon them within predefined policy,” Kramer said.

What Sets Aim Security’s Approach to AI Protection Apart

Aim’s platform covers everything from secure development lifecycle and AI posture management to production monitoring and usage control of both internal and third-party AI applications, Kramer said. Aim’s research capabilities give it a critical edge, with the company’s research team being among the first to expose novel and highly sophisticated AI attack techniques, according to Kramer.

“The advantage of Aim is twofold. One, its breadth,” Kramer said. “It really covers everything from secure development and posture management to production of internal applications and external applications to controlling the usage of third-party applications. There’s no other vendor that I know that has that breadth as well as depth.”

Traditionally, network security was concerned with traffic inspection and application and data protection, but Kramer said AI now introduces an entirely new class of communication and decision-making that requires a different kind of visibility and control. The network is the most effective control point for securing AI interactions since it captures conversations between users, apps and autonomous agents.

“That’s something that is unique to us, because we are the only ones that actually embed these three layers in the network,” Kramer said. “Our competitors have virtual appliances that run on the network, and thus essentially, AI is going to be another virtual appliance running on the network that taps the traffic and sends it somewhere.”

The network provides a centralized and privileged vantage point, capturing all traffic between users, apps and agents without the need to instrument each endpoint, Kramer said. Conversely, enforcement via endpoint agents or secure enterprise browsers requires deployment across potentially tens of thousands of endpoints, is difficult to maintain, and provides fragmented visibility, according to Kramer.

“If you want to listen to conversation between people, applications, and agents, and how all of these are communicating with each other, this is on the corporate network, and we are the network,” Kramer said. “We are embedded in the network. We can extract these conversations, so we are the best point to tap that and inject that new capability.”