Anthropic Partners with SpaceX to Expand AI Capacity

Anthropic will gain access to 300 megawatts of AI compute, doubling code rates.

Anthropic Partners with SpaceX to Expand AI Capacity
Photo Anthropic CEO Dario Amodei.

WASHINGTON, May 8, 2026 – Anthropic announced a major computing partnership with SpaceX on Wednesday to expand the infrastructure behind its Claude artificial intelligence platform.

Under the agreement, Anthropic will gain access to 300 megawatts of new artificial intelligence computing capacity through SpaceX’s Colossus 1 facility in Memphis, Tennessee, which houses more than 220,000 Nvidia GPUs.

It is also exploring partnerships with SpaceX to develop multiple gigawatts of orbital AI computing infrastructure.

Anthropic said the SpaceX partnership joins several previously announced infrastructure deals, including an agreement with Amazon for up to 5 gigawatts of AI infrastructure capacity, with nearly 1 gigawatt expected online by the end of 2026.

The company also announced a 5 gigawatt agreement with Google and Broadcom scheduled to begin coming online in 2027.

Anthropic separately announced a partnership with Microsoft and Nvidia that includes $30 billion in Azure computing capacity, alongside a $50 billion investment in American AI infrastructure with Fluidstack.

Morgan Stanley estimates capital expenditures by the five largest AI hyperscalers will exceed $800 billion this year and rise to $1.1 trillion next year, according to figures shared this week.

The partnership adds to a wave of AI infrastructure spending tied to rising demand for computing power, networking capacity, semiconductor supply chains, and large-scale data center construction.

The company said some future infrastructure expansion will occur internationally to support enterprise customers in regulated sectors including healthcare, financial services, and government that require regional data storage and compliance capabilities.

Popular Tags

#if @member /if