Elon Musk does not often admit he was wrong. But he has. Responding on X to a user who noted that SpaceX controls the computing power Anthropic depends on, and could in theory cut it off, Musk said he never would.
He went further, calling Anthropic the clear leader in AI and saying no rival had shipped a model as good as its Mythos and Fable systems. He expects a Mythos 2 before long.
The turn is stark. In September he wrote that winning was never a possible outcome for Anthropic. In February, after it raised $30bn at a $380bn valuation, he called its models "misanthropic and evil" and told it to fix them. Five months on, the same company is his benchmark for the field.
The praise is not free
Take the compliment at face value and it reads as rare humility. Look at the plumbing and a second motive appears. In May, Anthropic agreed to lease the entire output of xAI's Colossus 1 data centre near Memphis, about 300 megawatts, paying roughly $1.25bn a month through 2029. That is close to $40bn flowing to Musk's side of the table. He is not just admiring Anthropic. He is billing it.
A rival who pays you $40bn is a rival worth flattering. Musk's pledge not to weaponise that dependence costs him nothing and buys goodwill with a customer he needs. The admiration may be real. It is also good business.
The real target is Altman
The sharper reading sits one company over. Musk co-founded OpenAI in 2015 as a non-profit, left the board in 2018 after the others refused to hand him control, and has fought it since. He sued Altman and OpenAI in 2024, accusing them of abandoning the founding mission for private gain. He sought more than $150bn in damages, Altman's removal and the unwinding of the for-profit structure.
In May a jury threw the case out, finding Musk had waited too long to file. He called the verdict a "calendar technicality" and vowed to appeal. By July the feud was personal again. Musk branded Altman "Scam Altman" after Apple sued OpenAI, and Altman replied that the surest sign his new model led the field was that Musk was obsessed with him again.
Against that backdrop, crowning Anthropic the leader does double duty. It is a real judgement about the models. It is also a way to tell the market that the AI company that matters is not the one he is suing.
Both are racing to the same finish line
Timing sharpens the point. Both firms filed confidentially for stock market listings in June, within days of each other. Anthropic, valued at about $965bn in private markets, is pushing for a Nasdaq debut as early as October. It could be the first company to list at close to $1tn, and it expects its first profitable quarter, with around $559m in operating income on $10.9bn of revenue.
OpenAI, valued a little lower and still loss-making, is leaning toward 2027, wary after SpaceX's own listing spiked and then surrendered much of the gain. Altman is holding out for a $1tn price.
Here the two threads meet. Whoever lists first sets the benchmark the second is priced against. Anthropic going out ahead, blessed by Musk as the field's leader, helps fix the multiple bankers later apply to OpenAI. Musk talking up the rival he profits from, while running down the rival he is suing, shapes the terms on which his enemy will one day face public investors.
None of this proves the praise is hollow. Musk may well think Anthropic makes the best models right now. But "wowed by the model" and "at war with OpenAI" were never competing explanations. They are the same move. The compliment serves his balance sheet and his grudge at once, and it lands in the narrow window before both labs ask the market to price them.
