Despite calling Anthropic "evil" just three months ago, Elon Musk has leased his supercomputer to the AI lab in a deal that could generate up to $4 billion annually for SpaceX. This move is seen as a strategic play ahead of SpaceX's public roadshow next month, where the company is targeting a valuation between $1.75 trillion and $2 trillion.
The deal allows Musk to monetize the Colossus 1 supercomputer, which was initially built to train Grok, Musk's AI assistant. However, Grok has not been generating enough revenue to justify the compute capacity, while Anthropic has a significant user base but lacks sufficient compute power.
"He who controls the data center, really does control the application of artificial intelligence right now," said Andrew Moore, former head of Google Cloud AI.
Analysts suggest that this deal positions SpaceX as a competitor to major cloud providers like AWS, Microsoft Azure, and Google Cloud. By capturing the margins typically paid to these hyperscalers, SpaceX could justify a higher valuation during its IPO. However, skeptics argue that building one massive data center does not replicate the global footprint and reliability of established cloud providers.
Additionally, Musk has included a clause allowing SpaceX to reclaim the compute if Anthropic's AI engages in actions that harm humanity. This clause, if enforceable, gives Musk significant leverage over the AI lab.