NVIDIA DGX GB300 supercomputer goes online at Naval Postgraduate School

NVIDIA CEO Jensen Huang personally visited the Naval Postgraduate School (NPS) in Monterey, California, to commission a DGX GB300 system, one of the most powerful AI computing platforms available. The deployment puts frontier-scale AI compute directly in the hands of students, researchers, and faculty at the U.S. military's flagship graduate institution, enabling defense-relevant research in areas from autonomy to simulation and cyber.

Operationally, DGX GB300 represents NVIDIA's Grace-Blackwell rack-scale generation, and siting it at a military research university signals both a public-sector market and a national-security positioning for NVIDIA's top-tier hardware. Reuters characterized the move as a supercomputer donation to a U.S. military university, underscoring the goodwill and strategic-relationship dimension.

Competitively, the NPS deployment fits a broader NVIDIA push into sovereign and national AI infrastructure this week — including the world's first national AI factory in Japan, built with 27,500 Rubin GPUs and 13,750 Vera CPUs. Together these signal NVIDIA cementing its role as the default compute layer for governments and defense, a market less price-sensitive and more strategically sticky than commercial cloud. The context also sharpens the week's U.S.-China AI-sovereignty theme, as Washington weighs restrictions on Chinese models while American institutions stand up domestic frontier compute. Watch what research NPS publishes on the system and whether other defense and government institutions follow with GB300-class deployments.