Anthropic’s IPO Filing Puts AI Growth Under Wall Street’s Microscope

Anthropic has filed to go public. The company submitted a Form S-1 registration statement to the U.S. Securities and Exchange Commission on October 26, 2026, for an initial public offering of its Class A common stock, with plans to list on the Nasdaq Global Select Market under the ticker symbol “ANT.”

The 87-document submission, filed under the Securities Act of 1933, was accepted at 9:00:11 a.m. Eastern. Anthropic is incorporated in Delaware and lists its principal executive offices on Market Street in San Francisco, with Chief Executive Officer and co-founder Dario Amodei and President and co-founder Daniela Amodei named as agents for service.

The preliminary prospectus covers shares sold by Anthropic and selling stockholders. Anthropic will not receive proceeds from shares sold by those stockholders, and the filing makes clear that the offering is not fully priced: the number of shares and estimated price range remain blank. The document also warns that its information is incomplete and may change, because even billion-dollar IPOs apparently require a little suspense.

A staggering growth story with an even larger bill

Anthropic was founded in 2021 and commercially launched its Claude models in the first quarter of 2023. The prospectus describes Claude’s uptake as “the fastest enterprise adoption of any technology in history,” with revenue rising from $0 at commercial launch to $16.3 billion in the first half of 2026.

The filing reports more than 11,900 customers generating $100,000 or more in trailing-twelve-month revenue as of September 30, 2026. More than 1,900 customers generated at least $1 million, while approximately 250 generated $10 million or more; Net Revenue Retention reached approximately 900% under the company’s measure, which compares current-period revenue from qualified customers with the same customers’ revenue in the year-earlier period and excludes Claude Pro and Max subscribers.

Revenue grew 1,368% year over year for the three months ended June 30, 2026, but compute and infrastructure expenses grew 300% over that period. Anthropic reported trailing-twelve-month revenue per employee above $6.6 million as of September 30, 2026, compared with $1.4 million a year earlier.

The company achieved profitability in the second quarter of 2026, yet reported an $80.7 billion net loss driven in large part by noncash accounting adjustments. As of June 30, Anthropic had raised $141.1 billion in capital and held $112.4 billion in cash and cash equivalents plus $19.3 billion in short-term investments.

Those figures sit beneath a valuation that reached $965 billion in May, while the offering could approach a $2 trillion valuation. Anthropic has also committed more than $500 billion in infrastructure spending arrangements with major technology partners over the next decade, as the AI industry contemplates $2 trillion in expected annual global spending.

Compute is the strategy and the liability

Anthropic warns that developing and training cutting-edge models depends on access to “enormous and increasing quantities of specialized compute.” If it cannot secure enough capacity or direct what it has across its workloads, the company says it risks falling behind the technological frontier.

The filing shows future obligations of approximately $417 billion for compute, including data centers, the equipment inside them and cloud services. Those deals begin between 2026 and 2027 and run from under one year to up to 20 years; additional capacity scheduled for delivery between 2028 and 2031 could add $125 billion to $150 billion because pricing has not been finalized.

Anthropic also says forecasts for future compute use are “inherently uncertain.” The company has underestimated its requirements before, affecting product development and deployment plans. That is a polite way of saying the machine needs more machines—and nobody knows exactly how many.

Claude is distributed through joint go-to-market arrangements with Amazon Web Services. Anthropic says Claude’s capabilities could advance scientific research, especially drug discovery, where the model already helps researchers automate workloads and analyze large data sets; future AI agents could operate microscopes, liquid handlers and off-the-shelf robotic arms.

The prospectus also mentions Claude Fable 5 and Mythos 5, while expansion in 2026 has been driven by Claude Code. Anthropic is racing against OpenAI and Google, even as it faces criticism over safety, regulation and national security, along with two high-profile disputes with the U.S. government in recent months.

Wall Street will weigh those risks against the growth. CoreWeave and Nebius have lost over 30% of their value in October as credit risk rises across the data center sector, while concerns grow over whether AI spending can justify the infrastructure costs ahead.

Anthropic employees and early investors will be able to sell shares on the first trading day, with staggered lock-ups opening 27, 48, 62 and 76 days after the IPO. Goldman Sachs & Co. LLC, Morgan Stanley, J.P. Morgan, Citigroup, Barclays, Wells Fargo Securities, BofA Securities, Deutsche Bank Securities, RBC Capital Markets, UBS Investment Bank and other listed underwriters are attached to the offering, with Wilson Sonsini Goodrich & Rosati and Freshfields US LLP serving as counsel.

The offering also extends to Australia, Canada, certain member states of the European Economic Area, Japan, Switzerland and the United Kingdom. The paperwork is filed; the price is not. For now, Anthropic has given investors the thing they wanted most—and the bill they may want to ignore.

Stay connected via Google News

Follow us for the latest AI News