Elon Musk, the billionaire entrepreneur behind Tesla and SpaceX, has publicly praised artificial intelligence (AI) firm Anthropic, a significant shift from his previous criticisms. This comes as Anthropic has committed to a multi-billion-pound deal to host its AI models on infrastructure provided by SpaceX's xAI unit, raising questions among some observers about the long-term implications of such a partnership with a direct competitor.
The agreement, signed in May, sees Anthropic utilising the entire 300-megawatt output of xAI's Colossus 1 data centre near Memphis, Tennessee. This substantial commitment is set to generate approximately £31 billion ($40 billion) in revenue for SpaceX's xAI unit, with Anthropic agreeing to monthly payments of £980 million ($1.25 billion) through to May 2029. Google has also signed a similar deal, renting SpaceX infrastructure for £720 million ($920 million) per month until June 2029.
Musk addressed concerns that he might leverage his control over the hosting infrastructure to disadvantage Anthropic, stating that such actions are "not my style." He highlighted Tesla's 2014 decision to open its patents and Supercharger network to competitors as evidence of his approach. He also pointed to SpaceX launching competing satellite systems without price increases or unfair terms, and allowing even his "worst enemies" to use his social media platform.
This public endorsement marks a significant turnaround from Musk's September 2025 assertion that "Winning was never in the set of possible outcomes for Anthropic." He now acknowledges Anthropic as the current leader in AI, specifically praising their Mythos/Fable models and anticipating the imminent release of Mythos 2. However, some analysts recall Musk's past actions, such as suing OpenAI, which underscore his competitive nature.
Beyond Musk's assurances, contractual obligations would likely protect Anthropic from any sudden withdrawal of services. The immense financial benefits for SpaceX in maintaining this lucrative partnership, combined with the potential for its engineers to gain valuable insights into supporting rapidly growing AI infrastructure, also serve as strong incentives to honour the agreement. Yet, the proximity of hosting could theoretically offer SpaceX greater visibility into Anthropic's operations, a factor that AI companies, including Anthropic itself, have previously raised concerns about regarding 'AI distilling' practices.