Elon Musk used SpaceX's first earnings call as a public company on Tuesday to make a bold case for why the aerospace manufacturer is uniquely positioned to lead the AI infrastructure race: its rocket scientists.

The remarks came after SpaceX spent nearly $16 billion on artificial intelligence infrastructure in the second quarter alone. Musk defended the aggressive investment by arguing that the engineering discipline required to design reusable rockets and satellites translates directly to building AI data centers. He compared SpaceX's competitive advantage to "the New York Yankees going in and playing a Little League team."

Data centers, Musk said, are a "trivial problem" relative to the challenges SpaceX's engineers tackle daily. "We're taking a small amount of the expertise that we use for rockets and satellites, and applying that to scaling terrestrial data centers," he said, adding that even a fraction of the company's aerospace engineering experience yields "an amazing outcome" when directed toward AI infrastructure. The central engineering challenge in hyperscale AI facilities is not computing itself but managing the enormous power draw and heat output of densely packed GPU clusters — a problem where thermal, structural, and systems engineering intersect.

Musk did not identify any of the competitors he characterized as "Little League." SpaceX is a relative newcomer to the data center sector, an industry long dominated by major cloud providers including Amazon, Microsoft, and Alphabet-owned Google. All three reported record cloud revenue growth in their most recent earnings. A growing field of "neocloud" startups is also competing to build the computing capacity needed to power AI workloads, and incumbents have collectively committed hundreds of billions of dollars to AI infrastructure over multi-year horizons, making scale and speed of deployment critical differentiators.

The comments underscore SpaceX's rapid expansion beyond its core rocket business. In February, the company acquired xAI, an AI venture also owned by Musk, and has since directed billions toward AI infrastructure. Total companywide capital expenditure reached $18.4 billion in the second quarter, nearly double the $10 billion spent in the first quarter. Executives on Tuesday's call indicated that spending levels are expected to remain roughly at current levels over the next two quarters.

SpaceX reported AI revenue of $2.6 billion in the second quarter, a 213% sequential increase, and announced $14.1 billion in contracted cloud services agreements. Contracts with Anthropic and Google are among the deals driving the company's emerging cloud business.

Beyond engineering talent, a key attraction of SpaceX's data center operation is its inventory of Nvidia GPU chips. Musk told analysts that SpaceX will build its AI computing platform exclusively on Nvidia's Vera Rubin architecture, which he described as the industry's best AI computing system. "Our understanding with Nvidia is that we will receive a very significant percentage of their GPUs next year," Musk said. Nvidia's GPUs are the most widely used accelerators for training large AI models, and access to limited supply has become a competitive fault line across the industry, with cloud providers competing to offer customers the latest chip generations.

That confidence extends to the scale of compute capacity SpaceX expects to bring online. The company ended the quarter with 1.4 gigawatts of nameplate compute and told investors it expects to exceed 2 gigawatts by year-end. When pressed by an analyst, Musk went further, stating that SpaceX's "tentative target" is 20 gigawatts of power and cooling capacity online by the end of 2027. Even accounting for delays, he said, the company would likely reach "close to 15 gigawatts." For context, individual hyperscale data centers typically operate in the range of 100 to 500 megawatts, and securing grid connections and energy supply at multi-gigawatt scale is one of the most cited constraints on industry-wide AI capacity growth.

While SpaceX has previously discussed the concept of space-based data centers powered by solar energy, Musk's focus on Tuesday remained on terrestrial facilities.

"Rocket science is literally our daily business," Musk said, noting that engineers on every flight work to prevent a rocket from doing what it "desperately wants to do" — blow itself apart. Applying even a fraction of that rigor to data centers, he said, is "kind of ridiculous, frankly," given the comparative simplicity.

This story was originally featured on Fortune.com.