Elon Musk announced that SpaceX will exclusively use Nvidia graphics processing units (GPUs) for its data centers, citing Nvidia's technological superiority and advanced chip supply chain. The declaration, made during SpaceX's first earnings call as a publicly traded company on August 4, 2026, caught the semiconductor market off guard and dealt a significant blow to AMD, which had been positioning itself as a direct competitor to Nvidia's dominance in artificial intelligence infrastructure.

What Prompted Musk's Exclusive Nvidia Commitment?

Musk's decision centers on SpaceX's ambitious vision to build data centers in space. During the earnings call, SpaceX revealed plans to secure more than 2 gigawatts (GW) of computing infrastructure by the end of 2026, expanding to around 10 gigawatts by the end of 2027. The company intends to launch Starmind AI satellites, which are space-based data centers, starting in 2027, and will equip them with Nvidia's Vera Rubin NVL72 architecture.

Musk stated on his social media platform X that "SpaceX has decided to use only Nvidia GPUs. Because they are the best." In the conference call, he elaborated on the technical reasoning, explaining that the NVL72 computer design is "a far better design than the standard rack style." The NVL72 integrates seven chips, including Nvidia's Rubin GPU and Vera CPU, connected through NVLink communication technology.

Musk

Nvidia's space-accelerated computing platform, unveiled in March 2026, delivers up to 25 times improved artificial intelligence computing performance in space-based inference compared with the previous H100 GPU. The platform includes the Space-1 Vera Rubin Space Module, Nvidia IGX Thor for spacecraft data processing, and the Jetson Orin ultra-compact inference module.

How Does Nvidia's Advantage Extend Beyond Raw Performance?

Nvidia's competitive edge extends far beyond chip performance alone. The company benefits from several structural advantages that make it difficult for competitors like AMD to catch up quickly:

• CUDA Software Ecosystem: Nvidia's CUDA software is optimized specifically to run on Nvidia chips and has become the industry standard for AI development. This creates a powerful lock-in effect, as developers and companies have invested years building applications around CUDA.
• Full-Stack Infrastructure: Nvidia offers an integrated ecosystem that includes not just GPUs but also networking equipment and complementary hardware. Analysts note that Nvidia's "infrastructure full stack" approach gives it an advantage over AMD's more fragmented offerings.
• Supply Chain Control: Nvidia is expected to overtake Apple in 2026 to become Taiwan Semiconductor Manufacturing Company's (TSMC) largest customer. This means TSMC is prioritizing Nvidia chip production as the industry shifts from mobile computing to AI data centers.

AMD does operate the AMD Versal AI Edge series platform targeting the aerospace market, but analysts say SpaceX preferred Nvidia's latest GPU modules. AMD also collaborates with Blue Origin, which is regarded as a SpaceX rival, potentially making the partnership less attractive to Musk.

Why Couldn't Musk Build His Own Chips Instead?

Musk has been pushing to build Terafab, a facility to directly produce AI chips for Tesla and SpaceX. However, analysts say acquiring foundry capabilities comparable to TSMC, Samsung Electronics, or Intel in a short timeframe is extremely difficult. After Musk announced plans to build a factory producing 1 million chips per month, Nvidia CEO Jensen Huang took a skeptical stance at a TSMC event in November 2025.

"Building an advanced chip manufacturing facility is very difficult. Not only building the factory itself, but the engineering, science and craftsmanship are all very difficult," said Jensen Huang.

Jensen Huang, CEO at Nvidia

Given the formidable challenges of establishing independent chip manufacturing, Musk's decision to commit exclusively to Nvidia reflects a pragmatic recognition that controlling a powerful chip supply chain is beyond SpaceX's near-term capabilities.

What Does This Mean for AMD's AI Strategy?

The timing of Musk's announcement proved particularly damaging to AMD. On the same day SpaceX made its exclusive Nvidia commitment, AMD announced earnings that exceeded expectations. Yet AMD's stock price plunged following Musk's remarks, signaling that investors view the loss of SpaceX as a major setback.

AMD CEO Lisa Su responded carefully to the announcement, attempting to reassure investors about the company's long-term prospects. In a CNBC interview on August 5, 2026, Su stated that she had "deep respect for Elon Musk and everything he has achieved" and expressed her intention to "continue our partnership over the long term".

Lisa Su

"I have deep respect for Elon Musk and everything he has achieved, and I look forward to continuing our partnership over the long term," said Lisa Su.

Lisa Su, CEO at AMD

The contrast with Musk's May 2026 comments is striking. Just three months earlier, Musk had indicated that Tesla and SpaceX would likely continue purchasing chips from both Nvidia and AMD. The sudden shift to an exclusive Nvidia arrangement suggests that SpaceX's space data center ambitions require capabilities that Nvidia uniquely provides.

Was This Decision Really a Surprise?

Some analysts argue that Musk's exclusive Nvidia commitment was less a dramatic strategic pivot and more a formalization of existing practice. xAI, SpaceX's AI development subsidiary, has primarily used previous Nvidia GPU generations, including Hopper and Blackwell, to train its AI model Grok. There is no public evidence that xAI developed Grok using chips from AMD or other competitors.

According to IT media outlet Tom's Hardware, "Because xAI's supercomputers have already used Nvidia accelerators for years, the exclusive deal officially announced this time appears to be closer to a formality than a decision reached after much deliberation." This perspective suggests that SpaceX's reliance on Nvidia was established long before the public announcement, making the exclusive commitment a natural extension of existing infrastructure investments.

The announcement underscores a broader trend in AI infrastructure: companies building cutting-edge systems increasingly find themselves locked into Nvidia's ecosystem through a combination of superior performance, mature software tools, and supply chain advantages. For AMD and other competitors, breaking into this market requires not just competitive chips but also the ecosystem depth and supply reliability that Nvidia has spent years building.