What’s behind Anthropic’s $518 billion AI buildout: New filing reveals crucial details
Anthropic is warning potential investors in its IPO that advanced AI could pose an existential risk to humanity
In a major update, Anthropic said it anticipates spending at least $518 billion over a decade building AI infrastructure with six partners-a plan that ranks among the largest AI recorded AI development commitments.
The AI primarily said in the prospectus that about 80% of that sum is a firm commitment that requires payment regardless of usage. Anthropic is securing these commitments which are vital to unlocking capital for AI development as future demand for advanced AI systems outstrips supply due to processing limits imposed by the availability of computers.
The company said it plans to spend at least $111.1 billion with Alphabet's Google, $110 billion with Amazon and $31.4 billion with Microsoft subject to a seven-year infrastructure service commitment. Anthropic said it is working to build its own AI infrastructure shifting from a hybrid infrastructure transition toward dedicated data centers and compute provisioning.
It identified a key risk in its reliance on Amazon, Google and Microsoft with the companies controlling both the underlying infrastructure stack and the consumer-facing AI models. Furthermore, the company said the three tech giants occupy multiple roles as investors, cloud providers, competitors, thereby creating incentives that conflict with Anthropic’s interests.
