SUSE has shown the market new capabilities so that companies can adopt private AI solutions for enterprise environments, while maintaining operational control, security and governance of their systems. SUSE strengthens enterprise security for autonomous AI agents by integrating the new NVIDIA OpenShell reference architecture into SUSE AI Factory with NVIDIA.
This architecture combines NVIDIA Vera Rubin for model inference, NVIDIA Vera CPUs running OpenShell for agent execution, and NVIDIA BlueField-4 DPUs running NVIDIA Sentry, based on NVIDIA DOCA, for independent monitoring of the host system and policy enforcement directly on the silicon.
Controlled environment to run the agents
As a launch partner of the OpenShell reference architecture, SUSE contributes to the development of an open, industrial-scale architecture for creating, isolating, monitoring, and governing autonomous AI agents with enterprise-grade security. Using Kubernetes sandboxes and API gateways, along with OpenShell, SUSE AI Factory with NVIDIA provides a controlled environment that regulates agent execution without compromising maximum performance. Its main capabilities include:
- Active governance based on the zero trust model: Monitoring proxies continually inspect agent traffic, authorize access according to defined policies, and block unauthorized network requests to help prevent code or data extraction.
- Data sovereignty and regulatory compliance: Enables agent code to run within the corporate perimeter, with access policies that restrict the transfer of enterprise intellectual property and proprietary data sets outside of private infrastructure.
- Predictable cost model: Private deployment of models, using NVIDIA NIM microservices, eliminates token fees for third-party public APIs and allows infrastructure investment to be planned based on predictable capacity.
Available with SUSE AI Factory with NVIDIA, the latest version incorporates zero trust agent security through the NVIDIA OpenShell reference architecture, along with SUSE Linux Enterprise Server (SLES) for NVIDIA. The solution is designed to offer out-of-the-box support for NVIDIA’s next-generation silicon platforms.
Together, these innovations provide a unified, enterprise-ready foundation that enables developers to create production-ready AI applications and autonomous agents; to the platform engineering teams, deploy them; and govern enterprise IT departments on any type of infrastructure.
“As autonomous AI agents and advances in next-generation silicon platforms transform enterprise operations, AI sovereignty becomes a non-negotiable requirement,” said Dr. Thomas Di Giacomo, Chief Technology and Product Officer at SUSE. “By combining enterprise Linux designed for next-generation hardware with active security for agents, SUSE AI Factory with NVIDIA OpenShell enables IT teams to rapidly scale the use of AI while keeping proprietary data, system availability, and budgets under tight control.”
SLES infrastructure designed specifically for NVIDIA
SLES for NVIDIA enables enterprises to accelerate AI workloads with immediate access to the latest NVIDIA hardware innovations, including the NVIDIA Vera Rubin infrastructure, which forms the core of the NVIDIA OpenShell reference architecture.
As a well-established and reliable enterprise Linux platform, SLES for NVIDIA is designed to seamlessly scale to rack-scale architectures, connecting the rapid evolution of silicon innovations with stringent operational stability and security requirements.
Drive innovation in the public sector with Sovereign AI
According to Gartner, by 2027, 35% of countries will be powered by region-specific AI platforms that use proprietary contextual data. Building on its existing relationship with SUSE, the Ministry of Finance of the Mexican state of Guanajuato is leveraging SUSE AI Factory with NVIDIA to validate the capabilities of sovereign AI with the goal of building a cognitive government serving citizens.
- Cognitive Government Proof of Concept: The Secretariat is actively testing SUSE AI Factory with NVIDIA to validate AI-based smart public services and optimize internal workflows.
- Hardware and Workload Acceleration: The proof of concept evaluates the deployment of AI workloads on a high-performance GPU infrastructure, facilitating the transition from initial software testing to production deployment in enterprise environments.
- Sovereign infrastructure foundation: By integrating SUSE AI Factory with NVIDIA into its global strategy based on SLES and SUSE Multi-Linux Manager, Guanajuato maintains full data sovereignty, centralized management and regulatory compliance in both hybrid cloud environments and on-premises infrastructures.
These innovations provide a unified, enterprise-ready foundation that enables developers to build AI applications.
“The use of SUSE AI Factory with NVIDIA in our environment offers us a clear path to becoming a truly cognitive government,” says José Luis Huerta Santos, from the IT Planning and Architecture area of the Guanajuato Ministry of Finance. “Having a sovereign, open source foundation allows us to quickly evaluate AI workloads, while ensuring that citizen data, infrastructure availability and operational costs remain under our strict control.”
Bringing hybrid quantum intelligence to operational environments with Fujitsu
SUSE and Fujitsu (FSAS) collaborate to deliver a hybrid AI-powered quantum computing platform that integrates SUSE AI Factory with NVIDIA and NVIDIA CUDA-Q software on Fujitsu servers. The solution is designed to move hybrid quantum intelligence from research laboratories to enterprise data centers. This marks the first time that enterprise-grade hybrid quantum computing and NVIDIA AI Enterprise are offered as a unified, production-ready technology stack running on sovereign infrastructure.
“As Fujitsu expands its technology portfolio with our new quantum computer and Digital Annealer, this collaboration provides the secure infrastructure foundation that businesses need,” said Udo Wuertz, Chief Technology Officer (CTO) of Fsas Technologies EMEA at FSAS. “Running this technology stack on Fujitsu servers allows customers to make the most of their current hardware investments and directly connect quantum simulation to applications in production.”
