NVIDIA plans to manufacture its AI supercomputers entirely within the United States for the first time, marking a significant shift in the global technology supply chain.
In partnership with leading manufacturers such as TSMC, Foxconn, Wistron, Amkor, and SPIL, NVIDIA is commissioning more than a million square feet of manufacturing space.
Production of NVIDIA’s next-generation Blackwell chips has already begun at TSMC’s facility in Phoenix, Arizona. New supercomputer manufacturing plants are being built in Texas — Foxconn in Houston and Wistron in Dallas — with mass production expected to ramp up over the next 12 to 15 months.
NVIDIA’s move is projected to generate up to US$500 billion in AI infrastructure within the US over the next four years. The company expects this investment to create hundreds of thousands of jobs and contribute trillions of dollars to US economic security in the coming decades.
“Adding American manufacturing helps us better meet the incredible and growing demand for AI chips and supercomputers, strengthens our supply chain and boosts our resiliency,” said Jensen Huang, Founder and CEO of NVIDIA.
Where AI Supercomputers Are Currently Produced
Until now, the production of AI supercomputers and their critical components has been dominated by facilities in Asia, particularly Taiwan and China.
For example, NVIDIA’s Blackwell chips have been primarily manufactured at TSMC’s plants in Taiwan and, more recently, at its new site in Phoenix, Arizona.
The complex supply chain for AI hardware has relied on advanced manufacturing, packaging, assembly, and testing technologies concentrated in Asia.
The decision to shift significant manufacturing capacity to the US comes as both the Biden and Trump administrations have both pushed for greater domestic production of semiconductors and high-tech infrastructure, citing national security and economic resilience.
Recent trade tensions and tariff threats have further accelerated efforts to localise critical technology manufacturing.
NVIDIA’s US-based supercomputer factories will leverage advanced AI, robotics and digital twin technologies, including NVIDIA Omniverse for factory simulation and Isaac GR00T for manufacturing automation, to optimise operations. The initiative is expected to catalyse the development of AI factories or data centres purpose-built for AI workloads, across the country.
Entelechy Asia 