SpaceX's first batch of Starmind AI satellites will be launched ahead of schedule by the end of 2027, partnering with NVIDIA to accelerate the development of orbital data center pathways

According to Zhitong Finance APP, Elon Musk is pushing the AI race from Earth to space. On Monday, SpaceX (SPCX.US) CEO Musk announced on the X platform that the company's first batch of Starmind AI satellites equipped with NVIDIA (NVDA.US) chips will be launched in the fourth quarter of 2027, reaching "significant scale" by 2028. This timeline is ahead of the 2028 plan disclosed in the previous prospectus.

On the same day, NVIDIA officially announced that SpaceXAI will deploy its next-generation Vera CPU to accelerate Agentic AI workloads and send the optimized Vera Rubin NVL72 rack-level system into orbit as the computational core of the first generation of Starmind satellites. This collaboration marks a critical turning point in moving the world's first space AI data center project from concept to engineering implementation.

From "Ground AI Factory" to "Orbital Data Center": Vera Rubin's Space Debut

NVIDIA's Vera is the first CPU designed specifically for AI agents, featuring 88 NVIDIA-designed Olympus cores and equipped with LPDDR5X memory with a bandwidth of up to 1.2TB/s. In workloads such as Agentic AI, reinforcement learning, and data processing, task completion speed can be up to 1.8 times faster than x86 CPUs.

Ian Buck, NVIDIA's Vice President of Hyperscale and High-Performance Computing, stated: "Agentic AI requires a new type of computing system—not only to generate answers but also to take action. Vera empowers AI agents with the ability to act in real-time—executing code, processing data, and coordinating complex tasks. SpaceXAI is extending this architecture from ground AI factories to the next frontier of orbital computing."

Mike Nicolls, President of SpaceXAI, pointed out: "Vera provides us with the necessary CPU performance and memory bandwidth to handle a large amount of coordination scheduling, code, and data processing work, while allowing GPUs to focus on their core tasks. This means higher performance AI agents, with each watt of computation yielding more valuable work."

Starmind: The "Space Supercomputer" of One Million Satellites

Starmind is the orbital AI data center constellation project planned by SpaceX. In January 2026, SpaceX submitted an application to the Federal Communications Commission (FCC) to launch up to one million computing satellites, expanding AI computing power for the ground through deployment in low Earth orbit Each Starmind satellite is essentially an optimized Vera Rubin NVL72 rack-level system, equipped with NVIDIA Rubin GPUs and Vera CPUs. According to previous disclosures from SpaceX, a single satellite has a computing power of 120kW (peak 150kW), with the hardware codename AI1. High-speed data transmission between satellites is achieved through laser links, forming a distributed space AI supercomputer. The first batch of Starmind AI1 satellites will be based on the optimized Vera Rubin NVL72 rack-level system, marking the first time NVIDIA's accelerated computing platform is sent into space.

Musk stated on X that this data center existing in satellite form is "simpler, cheaper, denser, and lighter than traditional racks." Compared to ground facilities, space deployment does not require a large amount of land and has relatively lower heat dissipation pressure.

Computing Power Scale: Aggressive Expansion from 2GW to 10GW to 20GW

Musk also provided a more ambitious computing power expansion target: In early August, during a quarterly earnings call, Musk revealed that SpaceX plans to have over 2 gigawatts of AI computing power by the end of 2026, and "closer to 10 gigawatts rather than 5 gigawatts" by the end of 2027. SpaceX AI will also expand its ground AI infrastructure in Tennessee and Mississippi. The company plans to accumulate computing power through multiple projects, aiming to reach 20GW by the end of next year.

Musk also admitted that some projects may not be fully realized on time, but he expects to achieve about 15GW of online computing power by the end of 2027. He also revealed that SpaceX expects to secure a "very large proportion" of NVIDIA GPUs by 2027.

Trillion-Dollar Capital Expenditure and Controversy Over "Non-Mainstream" Path

The cost accounting for space data centers is under close scrutiny from Wall Street. Wood Mackenzie estimates that the cost of building a 1 gigawatt orbital data center is about $170 billion, more than three times that of traditional ground data centers. Morgan Stanley analyst Adam Jonas predicts that SpaceX's AI-related capital expenditures will reach $53 billion in 2026, with an additional $130 billion in 2027.

However, supporters argue that space data centers can avoid regulatory obstacles and community backlash faced by ground data centers. SpaceX is reducing launch costs through the reuse of its Starship rockets, promoting the economic viability of orbital computing. Musk previously explained that before choosing NVIDIA systems, various chip solutions were evaluated, ultimately determining that NVIDIA systems were the best choice.

From Exclusive Cooperation to Deep Binding: SpaceX "All in" on NVIDIA

This collaboration represents a comprehensive upgrade in the relationship between SpaceX and NVIDIA. On August 4, during SpaceX's first quarterly earnings call after going public, Musk announced that the company had decided to exclusively adopt NVIDIA technology for AI computing power construction. Musk clearly stated at the time: "We have established close cooperation and partnership with NVIDIA on multiple levels, and we choose to exclusively use NVIDIA." SpaceX believes that NVIDIA's Vera Rubin architecture is "the best AI computer currently available." The two companies announced on August 4th that they are jointly developing the computing payload for the Starmind AI1 satellite, with each satellite equipped with NVIDIA's Vera CPU and Rubin GPU. SpaceX has informed investors that its space server chips will exclusively use NVIDIA.

NVIDIA's previously disclosed 13F filing on August 14th showed that the company holds approximately $21 billion in shares of SpaceX. This deep binding relationship is extending from the capital level to the technical level.

NVIDIA's earnings report is imminent on Wednesday: The space narrative becomes the latest catalyst

The timing of this collaboration announcement coincides with NVIDIA's quarterly earnings report scheduled for Wednesday (August 26th). The narrative of space AI data centers is adding a new dimension to NVIDIA's AI infrastructure story.

NVIDIA stated that the Vera Rubin platform integrates computing, networking, and software into a unified architecture aimed at reducing the cost per token. As SpaceX AI advances towards gigawatt-level computing capacity, Vera Rubin provides a unified architectural foundation from ground-based AI factories to orbital computing.

In the orbital environment, constraints related to power consumption, thermal management, bandwidth, reliability, and physical integration are far more stringent than those in traditional data centers. The collaboration between NVIDIA and SpaceX signifies that its AI accelerated computing system is extending from Earth to near-Earth orbit