SpaceX may lease AI compute to Microsoft for $150B deal
SpaceX’s AI compute leasing unit discussed a potential $150 billion deal with Microsoft this summer, according to reports. The talks stem from Microsoft’s need for 3 gigawatts of capacity to address a shortfall, though no agreement has been confirmed. SpaceX’s compute leasing business—backed by its Colossus data centers in Memphis—already serves major AI labs like Anthropic ($1.25B/month) and…
Key points
- SpaceX discussed a potential $150B compute lease with Microsoft to cover a 3 GW shortfall
- Current clients include Anthropic ($1.25B/month), Google ($920M/month), and an unnamed tenant ($1.11B/month)
- SpaceX’s annualized compute lease revenue reached $41.1B in September 2026, with capacity goals of 2+ GW by year-end
A Microsoft deal would add to SpaceX’s growing list of tech tenants, alongside Anthropic and Google. However, the success of this business hinges on meeting aggressive capacity targets and securing large-scale contracts. The concentration of revenue in a few key clients—like the $1.11B/month deal with an unnamed customer—also poses risks if relationships falter. For now, the focus remains on whether SpaceX hits its 2026 capacity goals and whether Microsoft talks materialize into a signed agreement.
SpaceX discussed leasing AI computing capacity to Microsoft
cryptobriefing.com · 1 October 2026
Loading the full article…
This text was published by cryptobriefing.com and written by Diego Almada Lopez. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Hardware & Compute
All →- Meta launches Ray-Ban Meta Gen 3 in India with Ranveer Singh · 1 src
- Synopsys unveils Autopilot platform for AI-driven chip design agents · 4 src
- Deepseek releases open-source TileLang for Huawei Ascend chips · 4 src
- Alphabet to launch Google AI chips into orbit on SpaceX Falcon 9 · 2 src
- Nvidia-backed Nebius buys Inferize to cut AI inference GPU costs · 1 src
Comments
via GitHub Discussions