Anthropic eyes $1.4 trillion valuation in massive IPO plans

The scale of the offering

Anthropic is preparing to enter the public market with a filing that could mark one of the largest initial public offerings in United States history. According to a November 10 SEC filing, the company intends to list on the Nasdaq under the ticker ANT. The developer of the Claude AI model plans to offer 825 million Class A shares to the public.

The proposed price range for these shares is set between $70 and $78. At the highest end of that range, the public offering would generate approximately $64.35 billion in gross proceeds. It is important to note that the company would not receive the proceeds from the 23.65 million shares being sold by existing stockholders.

This move follows a period of significant uncertainty regarding how much the company is actually worth. Investors are currently reassessing whether the rapid growth in AI revenue is enough to support such a massive valuation. The proposed listing puts the company's fully diluted valuation at nearly $1.4 trillion.

A massive hardware backing

The IPO structure includes a significant private placement of nonvoting Class N shares. Seven technology companies have agreed to invest a total of $26 billion through this arrangement, provided the IPO successfully closes. This investment structure is distinct from the leveraged Anthropic ETFs that were proposed to provide exposure to stock movements rather than direct ownership.

The heavy hitters in this group include Google and Nvidia, who are each contributing $10 billion. Other major players include AMD with a $2.5 billion commitment and $1 billion contributions each from Broadcom, Micron, and Microsoft. TSMC has also committed $500 million to the private placement.

These investments highlight a shift in how the AI industry is being financed. The companies that manufacture and supply the necessary hardware are now becoming primary financial backers for the software companies that purchase their products. This creates a deep financial link between the infrastructure providers and the developers using that infrastructure.

Revenue and net losses

Anthropic is reporting accelerating demand for its enterprise services. The company reported $33.2 billion in revenue during the first nine months of 2026. This growth led to an annualized revenue run rate of approximately $81 billion in October, which was a $10 billion increase from the previous month.

However, the company's financial disclosures also show significant net losses. Anthropic reported a nine-month net loss of approximately $104 billion. It is important to clarify that much of this figure reflects noncash accounting adjustments tied to the company's rising private valuation rather than a direct cash operating loss.

Investors are looking closely at these numbers to determine the sustainability of the business model. While the revenue run rate is a positive indicator of sales pace, it is an estimate and not a reflection of revenue already earned over a full year. The company must prove it can turn this rapid growth into a stable, profitable public entity.

The cost of computing

One of the biggest hurdles for Anthropic's valuation is the sheer scale of its infrastructure obligations. The company disclosed approximately $900.9 billion in future purchase and lease commitments. These obligations are substantially tied to the computing capacity required to run large-scale AI models.

Previous analysis of the company's costs has shown that infrastructure spending can sometimes exceed revenue, even when customer growth is high. These commitments represent future contractual obligations that the company is bound to fulfill. They are not a reflection of cash that has already been spent by the firm.

The scale of these costs raises questions about the long-term efficiency of AI models. Investors are watching to see if improvements in model efficiency can eventually offset the rising costs of the hardware required to run them. The public market will be weighing these massive capital requirements against the actual profitability of the AI sector.