Anthropic Retracts Initial IPO Details Circulated to Banks Detailing Plans to Raise Nearly $70 Billion

Details regarding a potential initial public offering by artificial intelligence company Anthropic were briefly circulated to banks before being retracted late on Monday, according to emails reviewed by Reuters. A representative from Anthropic sent a copy of a securities filing that detailed plans to sell roughly 867 million shares at a price range of $70 to $80 per share. Shortly afterward, a follow-up email was sent stating that the details concerning pricing were inaccurate.

The second email noted that Anthropic would provide an updated filing at a later date. Anthropic declined to comment on the matter. The communication to bankers did not specify which particular details were erroneous, and it remains undetermined what was incorrect in the initial transmission.

The retracted filing outlined plans to sell 866.7 million Class A shares, which included a secondary offering component. Of that total, Anthropic itself intended to sell approximately 843.3 million shares, while existing shareholders planned to offer an additional 23.3 million shares. At the top of the stated price range, the IPO would have raised $69.3 billion.

In addition to the public share sale, the original document indicated that Anthropic had secured $26 billion in commitments from major technology corporations through a private placement. Companies slated to participate in the placement of Class N common stock at the IPO price included Alphabet’s Google, Nvidia, Advanced Micro Devices, Broadcom, Micron Technology, Microsoft, and TSMC. This private placement was structured to close immediately following the IPO.

Anthropic was expected to formally launch its roadshow on Tuesday, with pricing anticipated in the following week. The development arrives at a time when several companies in the AI infrastructure and data center sectors have delayed or withdrawn planned market listings. Market participants have cited rising interest rates, a broader selloff in shares and bonds of AI operators such as CoreWeave, and ongoing questions regarding the sustainability of high-level AI spending.

While earlier investor discussions floated a potential valuation of up to $2 trillion for Anthropic, expectations have since been adjusted downward, with $1.5 trillion viewed as the high end of the range. Founded in 2021 by former OpenAI executives including Dario Amodei and his sister Daniela Amodei, the company has developed the Claude family of generative AI models.

Anthropic’s growth has been closely tied to key partners like Google and Amazon, which act simultaneously as investors, computing power suppliers, distribution partners, and competitors. The company previously disclosed nearly $900 billion in computing commitments with partners including Amazon and Google, highlighting its dependence on extensive future capacity. An amended filing earlier in the month also revealed that Anthropic's quarterly growth rate has declined sharply, despite recording revenue of nearly $17 billion for the three months ended September 30.

"The retraction of preliminary IPO figures highlights the extreme volatility and scrutiny surrounding large-scale artificial intelligence market entries. While the involvement of major tech giants in private placements demonstrates strong sector confidence, fluctuating valuations and massive infrastructure commitments require companies to exercise rigorous financial precision. Founders and investors navigating capital markets must ensure complete accuracy in regulatory disclosures, especially when managing high-stakes offerings in capital-intensive industries." — Dr. Shishir Gupta, Founder & CEO, StartupLanes