Musk Pushes Terafab Chip Deadline Forward

Elon Musk says the major chip equipment makers are moving faster to help his new Texas factory meet its aggressive schedule.

The story began when Tesla and SpaceX announced a joint venture to build a vertically integrated semiconductor facility in Gibbons Creek, Texas. Intel joined the project in April 2026 to provide its 14A manufacturing process technology. The plan relies heavily on extreme ultraviolet lithography systems from ASML, the world's sole commercial supplier of these expensive machines. Other key partners include American firms like Applied Materials, Lam Research, and KLA, along with Japan's Tokyo Electron. The companies are also in talks with approximately ten South Korean manufacturers for specialized components and packaging equipment.

On October 20, Musk revealed during a live SpaceX earnings call that these suppliers are expediting shipments to support the project. He stated the team is targeting volume chip production at the primary facility by the third or fourth quarter of 2028. While he called the timeline "highly aspirational" with "big error bars," he expressed deep appreciation for the suppliers' efforts to accelerate deliveries. SpaceX also plans to produce its first experimental test chip at a separate research fabrication facility in Austin, Texas, around the second quarter of 2027. According to unofficial sources, the first EUV tool for the main plant was originally slated for mid-2027 delivery, though it is unclear if the current expedited schedule changes that date.

Musk also outlined the massive scale of the operation, claiming that once fully operational, the commercial value of Terafab’s output could rival the GDP of the 20th largest economy in the world. For context, Switzerland, currently in that position, has a GDP of roughly $1.14 trillion. The facility aims to manufacture one terawatt of AI computing power annually, handling logic, memory, and advanced packaging all in one location. This internal production is designed to supply chips for xAI clusters, Tesla’s self-driving vehicles, Optimus robots, and SpaceX’s orbital data centers, reducing reliance on external foundries like TSMC.

To achieve these goals, Musk revealed that SpaceX and xAI are aggressively procuring next-generation NVIDIA hardware. He plans to have a 440,000-GPU "Vera Rubin" training cluster online by the end of the second quarter of 2027. He expects his AI companies to operate 2.5 to 3 gigawatts of computing capacity by the end of 2026 and is confident they will reach a total of 13 gigawatts by the end of 2027. This trajectory far outpaces other major projects, such as the restructured 10 gigawatt Project Stargate or Meta’s 5 gigawatt Hyperion, which are both expected to reach full capacity in the 2030s.

Despite the optimism, experts remain skeptical of the one-terawatt ambition. Research firm Gartner estimates that total global AI-optimized data center capacity is only about 44 gigawatts as of late 2026. That figure represents just 4.4% of Terafab’s planned annual output. While global AI computing capacity has grown by roughly 3.3 to 3.4 times per year since 2022, according to EpochAI, the leap Terafab proposes is still historic in scale. The project represents a massive bet that internal manufacturing can solve the industry’s growing shortage of cutting-edge AI processors.

The next step is monitoring the delivery schedules of the critical EUV lithography systems and other front-end processing equipment. If the suppliers can maintain the expedited pace Musk described, the facility could begin volume production in late 2028. However, the tight capacity across the semiconductor supply chain remains a significant hurdle. The success of Terafab will depend on whether these partners can deliver the specialized machinery on time to support such an ambitious production ramp-up.