Elon Musk Declares Anthropic the AI Leader and Vows to Never Pull the Plug on Their Deal

Elon Musk Does an About-Face on Anthropic

Not long ago, Elon Musk was dismissing Anthropic as a company that could never win. Today, he's calling them the leader in artificial intelligence. The dramatic reversal came after users on X raised concerns about whether Anthropic was taking a serious risk by hosting its models on infrastructure controlled by a well-known rival.

When users suggested that Musk could simply pull the plug on Anthropic's server access at SpaceX whenever it suited him, Musk pushed back — firmly and publicly.

"I was clearly wrong about Anthropic," Musk wrote, referencing a September 2025 post in which he had declared that winning was "never in the set of possible outcomes" for the AI lab. That prediction aged poorly almost immediately. Even at the time he wrote it, Anthropic had already captured the largest share of the enterprise AI market.

A Billion-Dollar Partnership Redefines the Rivalry

The relationship between Musk's companies and Anthropic has evolved far beyond a social media spat. As of July 2026, Anthropic stands as one of SpaceX's biggest clients.

The partnership took shape in May when Anthropic signed a landmark deal to purchase 300 megawatts of computing power — the full capacity of xAI's Colossus 1 data center located outside Memphis, Tennessee. The agreement is valued at $1.25 billion per month and runs through May 2029, representing roughly $40 billion in total revenue for SpaceX's xAI division.

For context, Musk's xAI merged with SpaceX back in February, consolidating his AI and aerospace ventures under one roof. Google also entered a similar arrangement, agreeing to pay $920 million per month for access to SpaceX infrastructure through June 2029.

Musk Praises Anthropic's Mythos/Fable Model

In defending the partnership and his own trustworthiness as a business partner, Musk offered enthusiastic praise for Anthropic's latest work.

"They are obviously currently the leader in AI. No company has released a model as good as Mythos/Fable, and they will undoubtedly have Mythos 2 ready soon. And I would never cut them off in a way that hurt them badly, even as a competitor. That's not my style," he wrote.

To back up that claim, Musk pointed to several historical examples of his companies taking a hands-off approach with rivals. He cited Tesla's 2014 decision to refrain from filing patent lawsuits against anyone using its technology in good faith, as well as Tesla's choice to open its Supercharger network and charging port design to competing automakers.

"SpaceX launches competing satellite systems with no increase in price or use of unfair terms. Even my worst enemies can attack me on this platform," he added.

A Track Record That Isn't Entirely Clean

Of course, Musk's history with competitors is not without its contradictions. He has pursued legal action against OpenAI, a company with which he has a long and complicated history. So while his stated "style" leans toward fair play, the reality has occasionally looked different.

Why Anthropic Isn't Entirely at Musk's Mercy

Despite the inherent tension in relying on a rival for core infrastructure, Anthropic is not operating purely on good faith. Any sudden disruption to the arrangement would trigger significant contractual consequences for SpaceX. Beyond legal protections, the financial incentives for Musk to honor the deal are enormous.

There's also a subtler benefit for SpaceX. By hosting Anthropic's computing workloads, SpaceX engineers gain hands-on experience building and maintaining infrastructure for one of the world's most advanced AI systems — a form of institutional knowledge that carries real long-term value, similar to what Amazon's engineers gain through their relationship with Anthropic.

The Question of AI Distillation

One issue lurking beneath the surface is the practice of AI "distillation" — a technique in which one AI company creates fake accounts to query a competitor's model at scale, gradually reverse-engineering how it functions. During his legal dispute with OpenAI, Musk himself acknowledged the practice is real, noting that "generally AI companies distill other AI companies."

Earlier this year, Anthropic accused three Chinese AI developers of doing exactly this to its Claude model. The question of whether SpaceX might gain undue insight into Anthropic's systems simply by hosting its infrastructure is one that Anthropic and Google presumably have safeguards against — though the arrangement does provide SpaceX with a level of operational visibility that few competitors ever achieve.

The Bottom Line

For now, the partnership appears to offer Musk's company nearly all upside. SpaceX collects billions in monthly revenue, gains technical expertise, and earns credibility as a neutral infrastructure provider. What that picture looks like as the three-year contract matures — and as the competitive dynamics of AI continue to shift — remains an open question.