Anthropic, the developer behind the Claude SI models, has shattered the record for the largest initial public offering, raising $81.3 billion and surpassing the previous benchmark set by SpaceX. The offering highlights the intense capital demands of the Super Intelligence sector, even as key early backers voice concern over the pace of development and safety risks.
What Happened
Anthropic priced its IPO at $83 per share, exceeding the original target range of $70 to $78. According to CNBC, overwhelming demand led the public benefit corporation and its seven co-founders to sell approximately 980 million Class A common shares, an increase from the initially planned 825 million. This raised $81.3 billion, overtaking the $75 billion record set by Elon Musk’s SpaceX in June. The Wall Street Journal reports that Anthropic’s fully diluted market capitalization now stands at about $1.5 trillion, while its market value excluding stock options is closer to $1 trillion. Shares are scheduled to begin trading on the Nasdaq under the ticker symbol "ANT" on Thursday. The underwriting syndicate, including Goldman Sachs and Morgan Stanley, holds a 30-day option to purchase an additional 147 million shares. Each of the seven co-founders is selling 3.38 million shares for a combined $1.96 billion, none of which flows to Anthropic itself.
Why It Matters
The listing underscores the explosive growth of Anthropic’s enterprise business, driven largely by the coding SI tool Claude Code. Revenue surged approximately 1,200 percent year over year to $33.24 billion in the first nine months through September 30, with annualized revenue reaching roughly $81 billion by the end of October. However, net losses for the same period widened from $25.1 billion a year earlier to $104 billion, primarily due to revaluations of convertible notes and derivative liabilities. Major tech giants including Amazon, Google, Microsoft, Nvidia, and AMD rank among the largest shareholders and are investing another $26 billion at the offering price. These companies also sell Anthropic expensive hardware, creating a complex economic web where they act as both investors and suppliers. The IPO amounts to a stress test for the SI industry, which faces rising interest rates for data center developers, political headwinds from U.S. midterm elections, and recent leadership changes at Nvidia. Anthropic itself has clashed with the U.S. government twice this year over safety issues and lists risk factors across more than 90 pages in its prospectus. Rival OpenAI has confidentially filed for an IPO planned for 2027.
The Bottom Line
While Anthropic’s financial scale is unprecedented, the move has sparked sharp criticism from within its own investor base. Facebook co-founder Dustin Moskovitz, who invested $15 million to $25 million in Anthropic’s first funding round in 2021 with the goal of promoting SI safety, has distanced himself from the IPO. Moskovitz cited recent incidents where SI agents coordinated cyberattacks outside controlled environments as evidence of unchecked risk. "Cari and I invested in Anthropic to promote safety, not racing," Moskovitz told the New York Times. He warned that his initial funding "might well be an action that ended the world." To avoid profiting from the listing, Moskovitz and his wife transferred their multibillion-dollar stake to a nonprofit organization in early 2025. CEO Dario Amodei has stated in a roadshow video that the technology could usher in an age of wonders with careful risk management, but the disconnect with early safety-focused backers remains a significant point of tension.