Anthropic — the developer of the Claude family of large language models — is encountering difficulties securing sufficient cloud capacity from Amazon Web Services, Google, and Microsoft, which increasingly view Anthropic as a competitor rather than a customer. That information appears in an S-1 filing that was leaked to The Wall Street Journal.
Projected cloud spending and alternative infrastructure
The leaked filing adds detail to earlier reporting by Reuters and the Financial Times, including Anthropic’s estimate that it could spend up to $750 billion on cloud contracts, a figure that includes deals still under negotiation. Because the major cloud providers have not fully met its needs, Anthropic is leasing its own data centers and partnering with third‑party data center operators. The company notes concern that some of those third‑party operators have a “limited operating history,” which could make them less reliable than hyperscalers.
Financial picture: large losses and concentrated revenue
According to the filing, Anthropic’s heavy cloud-related spending occurs against an $80.7 billion net loss in the first half of the year. During the same period the company reported $16.3 billion in revenue, with two customers accounting for roughly one‑fifth of sales.
Liability risks beyond insurance coverage
Building on previously disclosed catastrophic risk warnings, the S-1 cautions that Anthropic could be held liable for damages caused by harmful AI behavior to an extent that would exceed available insurance coverage.
Recent revenue and IPO timing
The company told prospective investors last week that third‑quarter revenue has reached $16.8 billion, a 45% increase from the prior quarter. Anthropic is reportedly targeting a November initial public offering.
Response to requests for comment
Anthropic did not immediately respond to TechCrunch’s request for comment.
