Anthropic Lists in Record-Breaking IPO Valued at Over $1 Trillion
Summary: The AI firm prices shares at $83, targeting up to $117 billion in net proceeds. Founders retain majority voting control despite reporting over $100 billion in year-to-date losses.
Anthropic went public on November 19, 2026, executing one of the largest initial public offerings in history by both valuation and capital raised. The artificial intelligence company expects to offer its shares at a price of $83, a figure that values the firm at more than $1 trillion. This pricing represents an upsizing from the company’s original regulatory filings.
Beyond the primary listing, Anthropic is conducting a separate private placement of Class N non-voting shares to specific strategic partners. This secondary transaction is designed to raise an additional $26 billion. According to the prospectus, if underwriters fully exercise their option to purchase extra shares, the total net proceeds from the offering are predicted to reach approximately $117.0 billion. Despite the historic scale of the listing, Anthropic’s valuation at the time of the IPO remains lower than that of SpaceX.
Founder Control and Financial Performance
The company’s founders have structured their equity holdings to maintain decisive influence over corporate governance. Through an entity known as Founder LLC, along with special Class F and Class A shares that carry enhanced voting rights, the founders will maintain more than 50% of the votes in the firm. This dual-class structure allows them to retain operational control even though they own a smaller portion of the total economic equity.
Financially, Anthropic enters the public markets with substantial reported deficits. The company disclosed massive losses totaling over $100 billion for the year so far. Management stated that these figures are largely accounting-driven and related to the valuation of convertible notes. When specific adjustments are applied, the last two quarters have shown positive adjusted net income.
Revenue growth has outpaced compute spending in absolute terms, though the cost base is expanding rapidly. In the most recent quarter, Anthropic’s compute bill grew by approximately 77%, while revenue increased by 47%. The company has committed to spending tens of billions on infrastructure partnerships with firms including Amazon, Google, and Broadcom. Specifically, Anthropic has an agreement with SpaceX to utilize all compute capacity at the Colossus 1 data center, providing access to over 220,000 NVIDIA GPUs.
Risks and Regulatory Context
The company’s S-1 filing highlights several operational and regulatory risks. Anthropic disclosed that approximately 5-7% of its revenue for the year came from customers who were later found to be violating usage rules. The filing warned that such customers might evade controls or re-register under new accounts, meaning revenue from these sources could recur despite enforcement efforts. Concentration risk is also present, with the top five customers accounting for about 25% of revenue for the nine months ended September 30, 2026. Geographically, approximately 71% of revenue originated from the United States.
The prospectus also reflects recent changes in federal terminology. Following Executive Order 14434, issued by the US president on September 29, 2026, federal agencies are directed to use the terms “Super Intelligence” and “SI” instead of “Artificial Intelligence” and “AI.” Anthropic’s amended filing now uses these terms interchangeably to align with the new directive.
Leadership ties to previous AI ventures remain a point of interest. CEO Dario Amodei and President Daniela Amodei formerly held vice president roles in research and safety, respectively, at OpenAI. Additionally, the IPO has drawn attention to past investments by Sam Bankman-Fried, who used stolen FTX customer funds to invest in Anthropic. Former FTX executives Caroline Ellison and Nishad Singh held equity in the company but forfeited it after pleading guilty.