Elon Musk just announced one of the boldest infrastructure bets in tech history: a $55 billion semiconductor factory in Texas called Terafab, with potential to scale up to $119 billion. The facility represents Musk's answer to a critical problem facing every AI company right now: there simply aren't enough chips to go around. By building his own foundry, Musk's companies can control chip design, fabrication, and supply without waiting in line behind Apple, Google, and other tech giants competing for limited capacity from Taiwan Semiconductor Manufacturing Company (TSMC).

Why Is Musk Building His Own Chip Factory?

The core issue driving Terafab is supply scarcity. Right now, a handful of foundries manufacture the advanced chips that power AI systems, and demand far exceeds supply. Musk's ecosystem of companies faces this bottleneck acutely. Tesla needs custom silicon for Full Self-Driving and the Optimus humanoid robot. SpaceX requires chips for Starlink satellites and Falcon rocket avionics. And xAI, now merged into SpaceX following a major restructuring earlier this year, needs massive compute capacity to train Grok, the company's AI chatbot that competes directly with ChatGPT.

By building Terafab, Musk consolidates these competing demands under one roof. The facility would give his companies direct control over chip design, fabrication, and packaging, eliminating the supply delays that have historically slowed product launches. For context, the Terafab project is larger than TSMC's entire Arizona manufacturing facility.

What Are the Scale and Technical Specifications of Terafab?

The numbers behind Terafab are staggering, even by tech industry standards. SpaceX filed the plans on May 6, 2026, in Grimes County, Texas, and the project details reveal an ambition to reshape semiconductor manufacturing in the United States.

• Phase 1 Investment: $55 billion for the initial facility buildout
• Total Potential Buildout: Up to $119 billion across multiple phases
• Facility Size: Roughly 100 million square feet at full scale
• Power Requirements: Over 10 gigawatts of electricity at full operation
• Process Node: 2-nanometer chips, among the most advanced in the world
• Annual Output Target: 1 terawatt of compute per year

To put this in perspective, 1 terawatt of annual compute output would be enough to train multiple large language models simultaneously. The power requirements alone, exceeding 10 gigawatts, are equivalent to the electricity consumption of a small city. Ben Bajarin, a chip analyst at Creative Strategies, told CNBC that Musk is embarking on a "15-year strategy," meaning this is not a quick win but a multi-decade commitment to controlling compute infrastructure.

How Does Intel Fit Into Musk's Chip Strategy?

One of the most significant developments in the Terafab story is Intel's involvement. The American chipmaker joined the project in April 2026 as a strategic manufacturing partner, marking a major breakthrough for Intel's struggling foundry business. Terafab will use Intel's upcoming 14A process to produce chips, giving Intel its first major external customer commitment in years.

This partnership is mutually beneficial. Intel gains a massive customer with guaranteed demand, while Musk gains access to Intel's manufacturing expertise without building the entire fab infrastructure from scratch. The market reacted immediately; Intel stock had its best month in April 2026, more than doubling in value. For Musk, the partnership reduces risk by leveraging Intel's existing knowledge while bringing his own capital and scale to the table.

What Changed When SpaceX Acquired xAI?

Earlier in 2026, SpaceX acquired xAI, and the combined entity is now valued at $1.25 trillion. On X, Musk announced that "xAI will be dissolved as a separate company," and the merged entity would be called SpaceXAI. This restructuring is crucial to understanding Terafab's purpose. xAI's compute needs, including training Grok and running AI inference clusters, are now SpaceX's responsibility to solve.

Musk

The merger also explains Musk's push for AI data centers in space. SpaceX has the launch capacity to deploy satellites. xAI brings the AI workloads and expertise. Terafab supplies the chips. Together, they form a vertically integrated AI stack that spans from silicon fabrication to satellite deployment to model training. This integration gives Musk's companies a competitive advantage that no other company can easily replicate.

Steps to Monitor Terafab's Progress

• June 3 Public Hearing: Grimes County will vote on tax abatement incentives needed to move the project forward on schedule. If approved, construction can begin as planned. If rejected, the timeline slips significantly.
• SpaceX IPO in June 2026: SpaceX is planning to go public at a $1.75 trillion valuation, which would provide capital to fund Phase 1 of Terafab and reduce reliance on external financing.
• Intel 14A Milestones: Pilot production at Terafab targets late 2026, with full-scale manufacturing by 2027. Any delays in Intel's 14A process would cascade through the entire timeline.
• xAI Integration and Grok Development: Watch how SpaceXAI restructures Grok's compute infrastructure and whether the company prioritizes Grok training on Terafab chips once production begins.
• Competitor Responses: Google, Amazon, and Apple are all racing to design their own AI chips. Terafab may accelerate their own in-house chip manufacturing efforts.

What Are the Risks and Challenges Ahead?

While Terafab represents an ambitious vision, several significant hurdles remain before it becomes reality. A filing is not a guarantee. Semiconductor fabs of this complexity require years to permit, build, and qualify. The first phase alone costs $55 billion, more than the entire market value of most companies. Sustained funding will be a challenge, even with SpaceX's planned IPO.

Other uncertainties include which chips get manufactured first, since Tesla, SpaceX, and xAI have different technical requirements. The Intel partnership is also unproven at this volume; 14A is a new process node that has not been manufactured at scale. Additionally, Musk's other ventures could distract focus. He recently agreed to spend $60 billion acquiring AI startup Cursor, which could compete for capital and attention.

Perhaps most critically, if AI demand cools unexpectedly, building $119 billion of capacity that customers don't need would be catastrophic. The semiconductor industry is cyclical, and overbuilding capacity has bankrupted companies before.

What Does Terafab Mean for the Broader AI Race?

Terafab signals a fundamental shift in how the AI industry thinks about compute. Chips are now the bottleneck constraining AI progress, and the companies that control chip supply will dominate the next decade. NVIDIA is currently the world's most valuable company precisely because it controls the chips everyone needs. If Musk successfully builds Terafab, his companies become less dependent on NVIDIA, which would shift the entire power balance in AI hardware.

The project also puts pressure on other tech giants. Google, Amazon, and Microsoft are all racing to design their own AI chips to reduce their dependence on NVIDIA and external foundries. Terafab accelerates this trend toward vertical integration, where companies control every layer of their AI stack from silicon to software.

The next major checkpoint is the June 3 public hearing in Grimes County. If approved, Terafab moves forward on timeline. If not, skeptics will gain ammunition, and the project's viability will come into question. For now, Terafab is the clearest signal yet that Musk views AI chip manufacturing as essential to his long-term vision for Tesla, SpaceX, and xAI.