Tech • AI • Robotics • Game
Elon Musk's New Update on $119B Terafab, Biggest Building on Earth!
10/10TL;DR
Elon Musk’s proposed Texas Terafab is envisioned as a vertically integrated semiconductor complex costing up to $119 billion, built to secure chip supply for Tesla and SpaceX as AI demand outpaces global manufacturing capacity.
KEY POINTS
Why Musk wants Terafab
Musk has framed the project as a response to two risks: heavy US reliance on Taiwan for advanced chips and a broader shortage as AI expands across data centers, vehicles, humanoid robots and space systems. The premise is that future demand for compute could overwhelm today’s supply chain, making domestic, large-scale production strategically necessary.
A bid to unify the chip supply chain
The plan goes beyond designing chips in-house. Terafab is intended to combine logic, memory, advanced packaging, testing and parts of the manufacturing infrastructure in one US complex, reducing bottlenecks created when wafers, high-bandwidth memory and packaging capacity are spread across multiple countries and suppliers.
Extraordinary scale
The project has been described as a 100 million square foot facility, about 9.3 million square meters and roughly 10 times the size of Gigafactory Texas. Musk has said it could become the largest and most valuable building on Earth, with a long-term value exceeding $1 trillion if it reaches a target equivalent to 1 terawatt per year of silicon.
Construction activity in Texas
Work appears to be advancing at a site in Grimes County, Texas. The land has reportedly been cleared and leveled, with excavation, gravel placement and early foundation preparation under way. Reports also indicate the cleared area grew by 40% to 50% within weeks, alongside increased truck traffic and new staging zones.
Budget and phased buildout
The first phase is estimated at about $16.3 billion, while total investment could exceed $100 billion, with a widely cited figure of $119 billion for the full multi-phase project. That spending would cover not just buildings but the highly specialized equipment needed to run one of the most ambitious fabs ever proposed.
What chips it would make
The facility is expected to focus on two major product lines. One would produce AI5 and AI6 inference chips for Tesla autonomous driving systems and Optimus robots. The other, potentially larger line, would make D3 chips designed to withstand radiation for SpaceX satellites and orbital computing systems.
Target output and equipment bottlenecks
Long-term discussions around the project have included output near 1 million wafers per month and demand for potentially hundreds of billions of chips per year. But such scale would collide with a major industry constraint: access to EUV lithography tools, where ASML remains the only commercial supplier and global production is limited.
Hints of unconventional manufacturing technology
A large circular structure in concept designs has fueled speculation about an in-house free electron laser system. If adopted, such a technology could represent a radical attempt to boost lithography throughput beyond today’s conventional fab model, though the concept remains unclear and unproven at industrial scale.
Power needs on the scale of a small city
Terafab would require constant electricity for clean rooms, cooling, vacuum systems, packaging lines and data infrastructure. Project configurations have mentioned more than 40 natural gas turbines at around 50 MW each, implying nearly 2 GW of on-site generation, plus substations, transformers, storage and backup systems to avoid outages that could ruin wafers and idle production lines.
Water and industrial self-sufficiency
Semiconductor manufacturing also depends on vast volumes of ultra-pure water for repeated wafer cleaning. To support self-sufficiency, the complex would need large-scale treatment, recycling and distribution systems, making it less a single factory than an industrial city dedicated to turning silicon, power and water into AI compute.
The expertise gap
Capital alone cannot instantly create advanced chip manufacturing know-how. That is why Intel has emerged as a potentially important partner or knowledge source, bringing decades of experience in moving semiconductor processes from development into mass production. Even if construction and utilities are solved, operational expertise remains one of the project’s hardest challenges.
CONCLUSION
Terafab represents an attempt to secure the future chip supply for Tesla and SpaceX by concentrating nearly the entire semiconductor chain in one place. Its success would hinge not just on money and construction speed, but on scarce tools, reliable power, water infrastructure and world-class manufacturing expertise.
Explain this