Anthropic leaked IPO filing shows $42 billion loss

Anthropic was founded in 2021 by former OpenAI employees, including chief executive Dario Amodei, who left OpenAI after disagreements over its direction and governance. It has long positioned itself as a more safety-conscious alternative, aiming to win enterprise customers, and recently lobbied for an AI slowdown while several executives and engineers publicly warned that AI could lead to the death of all humanity. Even so, it launched a more powerful version of its Opus model last week.

The leaked numbers show how expensive that race has become. The filing discloses $7.33 billion spent on computing and infrastructure last year, up threefold from 2024, and $518 billion in future cloud, computing, and infrastructure obligations. Amazon and Google have invested billions and supply much of its cloud infrastructure, while Anthropic has also struck computing deals with SpaceX and smaller providers. It held $20.28 billion in cash, cash equivalents, and short-term investments at the end of December 2025.

The listing, after several delays, could occur after the November U.S. midterm elections and would offer one of the first direct ways for public investors to back a leading frontier AI lab. Whether that debut lands well appears to hinge on whether the rapid revenue growth — and the customer base behind it — proves durable enough to justify the cash burn the filing lays bare.