Check Point launches MHO 200 to secure high-volume AI network traffic

Scaling security for AI workloads

Check Point Software Technologies has introduced the next-generation Maestro Hyperscale Orchestrator (MHO) 200. This new hardware is designed to help organizations secure fast-growing, largely encrypted AI traffic within data centers. It aims to solve a common problem where companies have to choose between keeping their network fast or keeping it secure.

Many organizations currently face a dilemma when managing high-speed data. They often switch off advanced threat prevention features to maintain acceptable performance levels. Alternatively, they might purchase oversized hardware appliances that leave significant capacity idle during normal operations just to handle sudden spikes in throughput.

The MHO 200 addresses these issues by providing a way to scale threat prevention elastically. As PR Newswire reported, the goal is to allow neoclouds, telcos, and government bodies to build AI infrastructure without overprovisioning hardware. It ensures that security keeps pace with the volume and speed required by modern AI demands.

How the MHO 200 works

The Maestro MHO 200 is a hyperscale firewall orchestrator built for flexibility rather than being tied to a single chassis. It works by combining firewalls of different models and generations into one unified system. This allows the infrastructure to grow dynamically as the organization's demand for network capacity increases.

One of the key technical features of the MHO 200 is its ability to split the system into up to eight independent security groups. Each of these policy domains can have its own specific rules while still maintaining full threat prevention capabilities. This modularity helps manage complex environments where different types of traffic require different security protocols.

The system also incorporates Agentic Network Security Orchestration to move operations from manual intent to automation. This technology automates tasks like policy tuning, health monitoring, and compliance across the entire multi-firewall system. It allows security teams to manage a large, complex setup as a single object rather than managing each individual component separately.

Meeting the 400GbE standard

High-end data centers and AI factories are increasingly adopting 400GbE as the new standard for network connectivity. The MHO 200 includes new 400GbE ports to support this shift toward higher network throughput. This hardware is specifically designed to handle the intense requirements of training, inference, and agentic workloads.

These workloads drive significant amounts of east-west traffic, which is the communication that happens inside a data center. Because much of this traffic is encrypted, it requires constant inspection to remain safe. The MHO 200 provides 4.5 Tbps of threat prevention throughput per system, which is a fourfold increase over the previous MHO 175 model.

The device maintains a compact 1RU form factor while delivering 12.8 Tbps of total fabric capacity. This allows for high-density deployment in environments where space is limited. It also features live patching, which applies security updates on the fly without requiring any system reboots or downtime.

Who is affected by this launch

The launch is primarily aimed at organizations modernizing their data centers or building specialized infrastructure. This includes neoclouds, which are cloud providers built specifically to run AI workloads by offering GPU capacity as a service. It also targets colocation operators and GPU-as-a-service providers who need to manage high-volume traffic.

Government bodies building sovereign AI infrastructure are another key group for this technology. These entities require high-level security for private AI projects and must ensure that their data remains protected. The MHO 200 helps these organizations discover and secure AI applications, large language models, and Model Context Protocol servers directly from the firewall.

Telecommunications companies also stand to benefit by using the orchestrator to protect their backbone networks. By integrating with NVIDIA BlueField data processing units, the system can support private AI projects. This ensures that large-scale infrastructure providers can offer secure, high-speed services to their own clients and end-users.

The path toward AI security

The MHO 200 is the latest step in a broader strategy by Check Point to redefine network security for the agentic era. The company previously launched the AI Network Firewall to bring security directly into the firewall for inspecting AI traffic. This was followed by the introduction of Agentic Network Security Orchestration to automate network operations.

By extending these protections to the hyperscale level, Check Point is addressing the scale at which modern networks must be protected. The company recently ranked first in the Miercom 2026 Hybrid Mesh Network Security Benchmark with 99.8% security effectiveness. This marks the fourth consecutive year they have led that specific assessment.

Moving forward, the focus remains on a prevention-first approach to secure the AI transformation. The MHO 200 ensures that as AI continues to reshape networks, organizations do not have to sacrifice speed for safety. It provides a scalable foundation for companies to grow their AI capacity while maintaining constant uptime and resilience.