There are some new details about the upcoming initial public offering from Anthropic, the artificial intelligence start-up behind Claude. Reuters reported on Sept. 29 that the company is seeking a valuation of $2 trillion, despite posting a net loss of almost $42 billion in 2025. Anthropic also reports in the prospectus that its AI models could pose a “catastrophic or existential risk to humanity” and might even resist efforts to shut them down.
But there’s one item in the prospectus that I’m most interested in, and it has nothing to do with Anthropic’s valuation or fears about AI overreach.
Instead, I’m drawn to the company’s disclosure that it plans to spend $518 billion in the next decade on cloud computing and AI infrastructure to support its AI expansion. And the vast majority — 80% of that spending — is non-cancelable or requires Anthropic to pay even if it doesn’t use the services.
That’s a lot of guaranteed money going to some of the biggest cloud computing companies on…
