Anthropic is going public. On October 26, it publicly filed its S-1 with the SEC, with plans to list on the Nasdaq under the ticker ANT. The price range, share count and valuation are still blank.
The filing finally puts hard numbers behind the hype. Revenue hit $16.29 billion in the first six months of 2026, up from $1.16 billion a year earlier. The second quarter alone brought in $11.56 billion.
The $80.7 billion loss, explained
The headline loss for the half is $80.7 billion. But $72.4 billion of it is a non-cash charge, and most of it traces back to Amazon.
Amazon put much of its money into Anthropic as convertible notes, debt that turns into shares at the IPO. As Anthropic's valuation soared, those future shares became worth far more, and Anthropic has to book that rising value as a cost. Amazon's own quarterly filing valued the notes at $97.9 billion as of June 30. None of it left Anthropic's bank account.
Strip that out and the business looks very different. The half-year operating loss shrank to $693 million, from $2.1 billion a year earlier.
Amazon is the biggest winner
Here's an unusual detail: no venture firm owns 5% of Anthropic, the threshold for being named as a major shareholder. In most tech IPOs, lead VCs hold double-digit stakes. Anthropic is simply too big: at its last private valuation of $965 billion, even a 1% stake is worth nearly $10 billion.
Amazon is a different story. It has invested about $18 billion in total, and its notes can convert into up to 1.84 billion shares, capped at 12% of the company unless Amazon waives the cap. On top of that, Anthropic has committed to spend at least $110 billion on AWS over the next 10 years, and at least $114 billion with Google Cloud through 2033.
In other words, the companies selling Anthropic its computing power are also among its biggest owners. The S-1 itself flags the catch: some of those cloud partners build rival AI models.
The most unusual risk section on Wall Street
Roughly 80 of the filing's 261 main pages are risk factors, nearly double the 48 pages describing the business. Among them: advanced AI could pose "catastrophic or existential risks to humanity," and models could show self-preserving behavior, like trying to resist shutdown or acting in ways "resembling blackmail."
It fits Anthropic's safety-first pitch. It also means public investors are being asked to buy into a company that warns its own product could cause irreversible harm.
Why the timing works
Businesses keep choosing Claude. The Ramp AI Index, which tracks what US companies actually pay for, shows more businesses now pay for Anthropic than for OpenAI.
What's still missing is the price. Until Anthropic sets a range and a share count, nobody knows what it will be worth on day one, or how much it plans to raise.
Get stories like this in your inbox every week. Subscribe to The Week in Perspective.
Sources: Anthropic Form S-1 (SEC) · Amazon quarterly report (SEC) · Anthropic · Ramp AI Index