Anthropic Flags IPO Draft Pricing as Wrong After $69.3 Billion Email
Anthropic sent banks an IPO draft outlining up to $69.3 billion, then said pricing details were inaccurate, Reuters reported. An update is promised.

Anthropic sent banks a draft securities filing late on Nov. 9 that described an IPO raising up to $69.3 billion. A follow-up email then said the pricing details were inaccurate, Reuters reported on Nov. 10.
An Anthropic representative said the company would update the filing. The second email did not name the incorrect details, Reuters could not work out what was wrong, and Anthropic declined to comment. The figures in the draft are therefore unconfirmed, and it is unclear whether other terms will change.
What the retracted draft described
The first email laid out 866.7 million Class A shares at $70 to $80 per share, including a secondary offering, which is a sale by existing holders rather than by the company. Anthropic would sell about 843.3 million of those shares, and current shareholders would sell another 23.3 million. The $69.3 billion top figure adds the company and shareholder sales together.
The draft also showed $26 billion of investments lined up alongside the IPO. Google, Nvidia, Advanced Micro Devices, Broadcom, Micron Technology, Microsoft and TSMC had agreed to join a private placement of Anthropic’s Class N common stock at the IPO price, with that deal closing right after the offering. This placement is a separate financing from the Class A share sale.
Where the timetable and filing stand
A formal roadshow, the round of investor presentations before pricing, was expected on Nov. 10, with pricing set for the following week. Both dates are reported expectations, not a schedule the company has confirmed.
On the public record, Anthropic’s SEC submissions history listed an amended IPO registration filing dated Nov. 2 as its latest entry. No later entry contains the draft that was emailed to banks.
Valuation hopes and spending commitments
Reuters set the offering against a tough market for AI-related listings. Other AI infrastructure and data-center companies have delayed or withdrawn offerings, interest rates are rising, and shares and bonds of AI operators such as CoreWeave have sold off. Questions about whether AI spending can last have grown as well.
Investors had earlier talked about an Anthropic valuation of as much as $2 trillion. According to the report, expectations have since dropped, and $1.5 trillion is now seen as the high end.
Anthropic had nearly $900 billion in computing commitments with partners that include Amazon and Google. Those commitments cover future capacity, and some depend on financing outside Anthropic’s control. Amazon and Google also back the company financially while supplying computing capacity and distribution, even though their own AI work competes with Claude.
What this means for investors and the deal
Revenue was almost $17 billion for the three months ended Sept. 30, Reuters said, and the quarterly growth rate had slowed sharply. Against that backdrop, the size and terms of the offering matter for how Anthropic pays for future computing capacity.
Until the promised update arrives, the retracted draft shows how large the financing could be but not which pricing terms still stand. The updated filing, the roadshow and the pricing expected the following week are the next steps to watch.
Source: Unite.AI



