In short
- Elon Musk’s SpaceX/xAI signed a deal to supply Anthropic entry to the Colossus 1 AI supercomputer cluster.
- Anthropic stated the added compute energy will enhance capability for Claude Professional and Claude Max customers.
- The partnership pairs two AI firms which have taken sharply completely different positions on regulation and politics.
Elon Musk’s SpaceX—which merged with xAI earlier this yr—is teaming up with Anthropic to assist energy Claude, creating an uncommon partnership between firms which have discovered themselves on reverse sides of President Donald Trump’s administration.
The businesses introduced Wednesday that Anthropic will achieve entry to Colossus 1, xAI’s large AI supercomputer cluster constructed to coach and run superior AI fashions. Anthropic stated the settlement will give it entry to greater than 300 megawatts of further compute capability and assist enhance service for Claude Professional and Claude Max subscribers.
“Within the subsequent few days, we’ll be ramping up Claude inference on Colossus. Grateful to be partnering with SpaceX right here,” Anthropic co-founder Tom Brown wrote on X. “We’re going to want to maneuver quite a lot of atoms with a view to sustain with AI demand, and there is no person higher at rapidly transferring atoms (on or off planet Earth).”
In a follow-up publish on X, Musk stated conversations with senior Anthropic workers helped persuade him the corporate was approaching AI growth responsibly.
Similar right here.
By the use of background for many who care, I spent quite a lot of time final week with senior members of the Anthropic group to grasp what they do to make sure Claude is sweet for humanity and was impressed.
Everybody I met was extremely competent and cared a terrific deal about…
— Elon Musk (@elonmusk) Might 6, 2026
“Everybody I met was extremely competent and cared a terrific deal about doing the suitable factor,” Musk wrote. “Nobody set off my evil detector. As long as they interact in vital self-examination, Claude will most likely be good.”
Musk added that “SpaceXAI” had already moved coaching to Colossus 2, a a lot bigger next-generation AI supercomputer cluster designed to surpass Colossus 1’s roughly 220,000 Nvidia GPUs and help future Grok fashions.
The partnership comes as Musk and Anthropic CEO Dario Amodei have more and more represented completely different camps contained in the AI trade. Musk has aligned himself carefully with Trump and the U.S. Departments of Justice and Protection. On the identical time, Anthropic has positioned itself as one of many trade’s strongest advocates for accountable AI security requirements.
That stance not too long ago put the corporate in battle with the Trump administration and the Pentagon after Anthropic refused calls for to permit unrestricted navy use of Claude. The dispute centered on safeguards blocking the mannequin from getting used for mass home surveillance or totally autonomous weapons. Trump later ordered federal companies to part out Anthropic’s know-how, whereas Pentagon officers argued navy contractors should allow “any lawful use” of AI programs utilized by the federal government.
The settlement underscores how the race for computing energy is reshaping alliances throughout Silicon Valley. In February, Elon Musk folded xAI into SpaceX, arguing that electrical energy demand and cooling necessities are limiting the expansion of Earth-based AI knowledge facilities. Musk clarified on X on Wednesday that the xAI entity will probably be dissolved, and that a part of the enterprise will probably be referred to as “SpaceXAI” going ahead.
Additionally in March, Musk accused Anthropic of being “misanthropic” after the Claude developer raised $30 billion in funding at a $380 billion post-money valuation.
Regardless of his earlier harsh phrases, Anthropic stated the SpaceX deal provides to a rising record of infrastructure partnerships that already consists of Amazon, Google, Broadcom, Microsoft, and Nvidia.
“As a part of this settlement, we’ve got additionally expressed curiosity in partnering with SpaceX to develop a number of gigawatts of orbital AI compute capability,” Anthropic stated in a press release.
Whereas orbital knowledge facilities stay experimental and face main engineering hurdles, SpaceX stated terrestrial energy, cooling infrastructure, and accessible land could not scale quick sufficient for future AI programs, driving curiosity in space-based compute as a possible supply of bigger and extra sustainable vitality provides.
“If engineering challenges will be overcome, space-based compute affords near-limitless sustainable energy with much less affect on Earth,” the corporate stated.
Each day Debrief Publication
Begin every single day with the highest information tales proper now, plus authentic options, a podcast, movies and extra.

