Terafab Expansion: Musk promises accelerated chip output by 2028
Tesla and SpaceX aim for mass chip production by late 2028, backed by expedited equipment deliveries from major semiconductor suppliers.
By Ama Boateng ·
For consumers and tech investors, the pace of artificial intelligence hardware development is set to accelerate. Elon Musk announced yesterday that his joint venture with Tesla and SpaceX, known as Terafab, is targeting volume chip production by the third or fourth quarter of 2028. While Musk described this timeline as highly aspirational with significant uncertainty, the announcement signals a major push to increase domestic manufacturing of advanced processors.
The CEO revealed the details during a live broadcast of SpaceX’s Q3 2026 earnings call on X. He stated that key suppliers, including ASML, are moving equipment faster to support the project’s aggressive schedule. The primary facility is located in Gibbons Creek, Texas. A smaller research fab in Austin, Texas, plans to produce its first experimental test chip around the second quarter of 2027.
Musk expressed gratitude to ASML and other partners for accelerating deliveries. He emphasized that these are very aggressive dates. The Terafab project is a vertically integrated chip factory designed to manufacture 1 terawatt of AI computing power annually. It will produce logic, memory, and advanced packaging in a single location. This output is intended to supply chips for xAI clusters, self-driving vehicles, Optimus robots, and SpaceX’s orbital data centers, reducing reliance on external foundries.
The facility will utilize extreme ultraviolet lithography systems based on Intel’s 14A manufacturing node. Intel joined the venture in April 2026 to co-develop this technology. ASML, the sole commercial supplier of the required EUV tools, is a critical partner. Unofficial sources indicate the first EUV tool was originally slated for mid-2027 delivery, though it is unclear if that date has been moved up. Other suppliers include Applied Materials, Lam Research, KLA Corporation, and Tokyo Electron.
Beyond the fab, Musk outlined broader computing goals. He expects his AI companies to have 2.5 to 3 gigawatts of capacity by the end of 2026. He is confident about adding 10 gigawatts in 2027, which would total 13 gigawatts. To achieve this, SpaceX and xAI are procuring next-generation NVIDIA hardware, with a plan to bring a 440,000 GPU Vera Rubin cluster online by the end of Q2 2027.
Experts remain cautious about the 1 terawatt target. Gartner estimates total global AI data center capacity at roughly 44 gigawatts as of late 2026. This figure represents only 4.4% of Terafab’s planned annual output. Despite the ambitious scale, the project aims to secure a stable supply of cutting-edge processors for the growing AI industry.
The next step involves coordinating with about ten South Korean equipment makers, including Hanmi Semiconductor and HPSP. These firms provide specialized packaging, annealing, and piping components. The success of the 2028 deadline will depend on how effectively these suppliers manage the tight capacity across the global semiconductor chain.
Source: Tom's Hardware
