Nvidia to invest $500 bn to build AI servers, working with TSMC, Foxconn, Wistron

Nvidia plans to invest up to $500 billion over the next four years to build AI servers in the U.S., working with partners like TSMC, Foxconn, and Wistron.

NVIDIA Campus

The initiative includes the production of Blackwell AI chips at TSMC’s Arizona plant and supercomputer manufacturing facilities in Texas, expected to ramp up within 12 to 15 months, Reuters news report said.

The move aligns Nvidia with a broader tech industry trend of reshoring manufacturing amid pressure from the Trump administration and looming tariffs. Most of Nvidia’s chips are currently made in Taiwan, and the company is responding to both geopolitical risks and tariff threats by diversifying its supply chain.

Nvidia CEO Jensen Huang emphasized that U.S.-based manufacturing will enhance supply chain resilience, meet surging AI chip demand, and create hundreds of thousands of jobs. While some analysts question the $500 billion figure as potentially inflated, the company’s shift reflects a longer-term strategy to reduce reliance on Asia and strengthen its U.S. footprint in response to political and economic pressures.

NVIDIA has partnered with major manufacturers to produce AI supercomputers entirely in the U.S. for the first time.

Over a million square feet of manufacturing space has been commissioned in Arizona and Texas for the production of Blackwell chips and AI systems.

Chip production has begun at TSMC facilities in Phoenix, while supercomputer factories are being established with Foxconn in Houston and Wistron in Dallas, with mass production expected within 12–15 months.

NVIDIA is also collaborating with Amkor and SPIL for packaging and testing in Arizona. Within four years, the company aims to generate up to half a trillion dollars in U.S.-based AI infrastructure, enhancing supply chain resilience and deepening partnerships with global leaders. These AI supercomputers will power next-generation data centers known as AI factories, with tens of gigawatt-scale sites anticipated.

U.S. manufacturing of these systems is projected to create hundreds of thousands of jobs and significantly impact economic and technological security. NVIDIA will apply its own AI, robotics, and digital twin platforms like Omniverse and Isaac GR00T to design and automate its facilities.

TelecomLead.com News Desk

0 0 votes
Article Rating
Subscribe
Notify of
guest
0 Comments
Oldest
Newest Most Voted

Latest

More like this
Related

Global Semiconductor Market to Hit $1.65 Trillion in 2026 as AI Boom Drives Memory Growth of 305 Percent

The global semiconductor industry is heading toward a historic...

Semiconductor Revenue Forecast to Surge 94% in 2026 as AI Drives DRAM, NAND and HBM Demand

Global semiconductor revenue is heading for extraordinary growth in...

Apple, Samsung, Google and UNISOC Gain as Qualcomm, MediaTek SoC Shipments Fall Over 25%; GenAI Chips Grow 24%

Global smartphone system-on-chip (SoC) shipments dropped 15 percent during...

Qualcomm Q3 2026: Revenue Hits $9.9 bn as Automotive Jumps 61%, IoT Reaches $1.83 bn and AI Diversification Accelerates

Qualcomm reported fiscal third-quarter 2026 revenue of $9.9 billion,...