How Foxconn Is Becoming Nvidia’s AI Factory for the World

AI Demand Fuels Foxconn’s Record Financial Performance

Foxconn arrived at Computex 2026 in its strongest financial position ever, a direct result of the AI infrastructure buildout. The company posted Q1 2026 consolidated revenue of NT$2.11 trillion (US$66.95 billion), a roughly 30% year-over-year increase . March alone set a single-month record at NT$803.7 billion (US$25.1 billion), up 45.6% from the prior year .

Foxconn now holds approximately 40% of the global AI server market and expects full-year AI server shipments to more than double in 2026 . Chairman Young Liu has described AI server rack shipments growing at a “high double-digit” pace sequentially, with Q2 2026 AI server revenue projected to nearly double both on a quarterly and annual basis .

The company’s cloud and AI subsidiary, Foxconn Industrial Internet, amplified this picture. It reported Q1 2026 revenue of RMB 251.1 billion, a 56.5% year-over-year surge, with cloud-service-provider AI server revenue more than tripling and AI GPU cabinet shipments growing 3.8-fold .

The $1.4 Billion Supercomputing Center in Taiwan

Nvidia and Foxconn are jointly building a $1.4 billion supercomputing center in Kaohsiung, Taiwan, which is scheduled for completion by the first half of 2026 . The 27-megawatt facility is designed to be Taiwan’s largest advanced GPU cluster and Asia’s first data center powered by Nvidia’s Blackwell GB300 chips—the immediate predecessor to the Vera Rubin platform .

First announced at Computex 2024, the Kaohsiung site serves as a physical anchor for the partnership. It will house 10,000 Nvidia Blackwell GPUs provided by Foxconn subsidiary Big Innovation Company, delivering computing infrastructure to researchers, startups, and industries across Taiwan .

An AI Factory in Texas Powered by Humanoid Robots

The partnership’s most futuristic project is a new AI server manufacturing plant in Houston, Texas. Foxconn and Nvidia are in advanced discussions to deploy humanoid robots at this facility, which will assemble Nvidia’s GB300 AI servers . If finalized, it would represent the first time Nvidia’s products are manufactured with humanoid robot assistance, and Foxconn’s first AI server production line to use them .

Foxconn aims to integrate robots powered by Nvidia’s Isaac GR00T N model, targeting initial deployment as early as Q1 2026 . The robots are expected to perform tasks such as pick-and-place operations, cable insertion, and basic product assembly . Foxconn has described the Houston site as a “world-leading benchmark AI smart factory” . The plant is part of Nvidia’s broader strategy to manufacture AI supercomputers entirely within the United States, alongside a separate facility in Dallas with Wistron .

The Deeper Meaning of the Partnership

Foxconn currently produces roughly 1,000 AI server racks per week for Nvidia and is scaling that capacity throughout 2026 . But the Computex 2026 meeting made clear that this relationship transcends a traditional contract manufacturing deal.

The alliance now spans three critical domains: high-volume production of Nvidia’s most advanced AI hardware in Taiwan, a flagship supercomputing facility in Kaohsiung that serves as a national AI resource, and a next-generation automated factory in Texas that embodies Nvidia’s “AI factory” vision. Foxconn is leveraging Nvidia’s Omniverse digital twin technology to simulate and optimize these production lines virtually before physical deployment .

As Jensen Huang has described it, the Vera Rubin rollout could be “the largest product rollout Taiwan’s electronics industry has ever seen” . The meeting between Huang and Liu confirmed that Foxconn is not just along for the ride—it is the vehicle.