< Back to all clusters
[TECHNOLOGY] · United States · 36 sources

started · updated

SpaceX and NVIDIA partner to launch orbital AI supercomputers by 2027

SpaceX and NVIDIA have announced a partnership to develop and deploy orbital AI computing infrastructure. The collaboration centers on a space-optimized version of NVIDIA’s Vera Rubin NVL72 system, which integrates 72 Rubin GPUs and 36 Vera CPUs. This hardware is designed to power SpaceX’s next generation of agentic AI applications, including the Grok chatbot.

SpaceX plans to launch its first-generation Starmind AI satellites equipped with this technology in the fourth quarter of 2027, with a goal of achieving significant scale in 2028. The company’s computing ambitions are massive, aiming to reach over 2 gigawatts of capacity by the end of 2026 and nearly 10 gigawatts by the end of 2027.

Elon Musk noted that the orbital design is significantly simpler, denser, lighter, and more cost-effective than traditional terrestrial data center racks. While the project offers potential advantages in bypassing terrestrial regulatory and land-use constraints, experts highlight significant technical challenges, particularly regarding heat dissipation in a vacuum environment. To support these massive workloads, SpaceX is also expanding its terrestrial AI infrastructure in states like Tennessee, Mississippi, and Texas.

Entities

Elon Musk · Japan · Ministry of Economy, Trade and Industry · Noetra Corp. · Nvidia · OpenAI · SpaceX · SpaceXAI · Starmind AI · Vera Rubin

Claims

What the coverage asserts, and how many sources carry each claim.

Sources