Sovereign AI infrastructure is now a board level concern. Rhys Oxenham, VP and GM of AI at SUSE, joins Alan Shimel to explain why teams must own where, when and how their intelligence runs. Furthermore, Rhys unpacks the SUSE AI Factory strategy, the new NVIDIA embed and the fresh Vultr partnership.

About Rhys Oxenham

Rhys studied computer science before joining SUSE, and today he leads the AI business unit. Consequently, he brings deep infrastructure experience to a company that already serves mission critical Linux, Rancher and edge deployments worldwide.

Two lenses on sovereign AI infrastructure

Alan and Rhys explore the two lenses SUSE uses. Infrastructure for AI is about running generative AI inference and machine learning operations on a hardened stack. Meanwhile, AI for infrastructure is about applying AI to CVE patching, security posture and day to day operations.

As a result, SUSE plays across both sides of the sovereign AI infrastructure market. In addition, 30 years of Linux, Rancher and edge experience compound into a strong foundation for modern AI workloads.

SUSE AI Factory with NVIDIA and Vultr

Rhys then walks through the SUSE AI Factory. Therefore, private enterprise AI becomes an assembly line experience across private cloud, public cloud and the edge. The NVIDIA embed layers NVIDIA AI Enterprise, NIMs, Nemotron toolkits and agentic frameworks into that same experience with a single support relationship.

The Vultr partnership then extends the deployment story across 32 global data centers with pay as you go economics. Explore more AI coverage and the latest TechStrong TV interviews.

Why sovereign AI infrastructure matters now

Alan and Rhys close on open source momentum. Open weight releases like Kimi K3 and Mistral give customers operational resiliency, predictable costs and a way out of vendor lock in. Consequently, sovereign AI infrastructure lets teams choose where the model runs, avoid export control shocks and never lose access to their own intelligence. Learn more at suse.com.