The deal now has a price range
Anthropic has put formal terms around one of the most closely watched public offerings of the AI boom. In an amended registration statement filed with the U.S. Securities and Exchange Commission on November 10, 2026, the company set an estimated IPO price range of $70 to $78 per Class A share and listed 825,000,000 Class A shares in the offering . The filing says Anthropic has applied to list on the Nasdaq Global Select Market under the ticker symbol “ANT” .
Those terms explain why the headline number has settled around $1.3 trillion, even though different outlets are using different share-count assumptions. AP described the transaction as an effort to go public at an approximately $1.3 trillion valuation, while Axios noted that the fully diluted value at the top of the range would be around $1.4 trillion . The gap is not just semantics: IPO valuation can be expressed using shares outstanding after the offering, shares included in concurrent private placements, underwriter options, employee awards and other instruments that may become stock.
At the offering midpoint of $74, the 825 million-share sale implies roughly $61 billion of gross stock sold in the IPO itself. At the top of the range, that same block would total $64.35 billion. The SEC filing also shows that Anthropic expects a concurrent private placement and estimates net proceeds from the IPO plus that placement of about $84.6 billion at the midpoint, or about $93.7 billion if underwriters fully exercise their option to buy additional shares .
Why this IPO matters beyond Anthropic
This is not simply another technology listing. Anthropic is one of the core frontier-model companies, best known for Claude, and the offering would give public-market investors a rare direct way to buy into a major AI lab rather than a diversified cloud, advertising or hardware company. AP cited analyst Dan Ives’ view that the deal would mark the first standalone public-market price for a U.S. frontier AI lab without being wrapped inside another large business line .
That standalone price is the real test. Investors have spent the past four years valuing AI labs on a mix of revenue growth, model quality, compute access, enterprise adoption and strategic importance. Anthropic’s IPO asks public buyers to attach a market value to those ingredients in real time. A successful debut would strengthen the argument that foundation-model companies can support mega-cap valuations as independent public companies. A weak debut would raise a harder question: whether private AI valuations have outrun what liquid public markets are willing to absorb.
The company’s own numbers make the debate sharper. AP reported, citing a prior regulatory filing, that Anthropic’s revenue reached more than $16 billion in the first half of 2026, about 14 times the level from the same period a year earlier, while the company still had not turned profitable . That combination—explosive growth and continuing losses—is familiar in technology IPOs, but the scale is not. Investors are being asked to price a company with unprecedented demand signals and unprecedented capital needs.
The valuation is large, but the structure is just as important
The amended filing highlights a corporate and capital structure that deserves as much attention as the headline valuation. Anthropic says it is a Delaware public benefit corporation, a legal form that is meant to align the company with a broader public-benefit purpose as well as shareholder interests . The filing also describes a multi-class share structure, including Class A, Class N, Class T, Class E and Class F stock .
The most consequential governance feature is the Class F share. According to the filing, the Class F common stock will carry votes equal to 50.1% of the total voting power on general stockholder matters, including the election of directors other than certain Class T directors . In plain English, public investors may get economic exposure to Anthropic, but they will not get ordinary control. That is common in founder-led technology listings, but Anthropic’s version is especially notable because the company is positioning itself not just as a software vendor, but as a developer of systems with broad social, economic and safety implications.
The filing also shows that Class N shares have no general voting rights while held as Class N stock, but may dilute the voting power of Class A holders when converted into Class A shares after qualifying transfers . Amazon and Google are listed in the filing as major holders of Class N stock immediately before and after the offering and concurrent private placement . That matters because Anthropic’s future will be watched not only through the lens of public shareholders, but also through its relationships with the largest cloud and AI infrastructure players.
Why the proceeds matter: compute, infrastructure and talent
Anthropic’s use-of-proceeds language is broad but revealing. The company says the principal purposes of the offering and concurrent private placement are to increase capitalization and financial flexibility, create a public market for Class A common stock, enable an orderly distribution of shares for selling stockholders and give the company and its stockholders access to public equity markets .
It also says it intends to use proceeds for general corporate purposes, including working capital, operating expenses, capital expenditures, compute and infrastructure investments, and hiring and retention of highly skilled personnel . That list captures the economic reality of frontier AI: model development is not just a research challenge, but a financing challenge. Training and serving advanced models requires chips, data centers, power, networking, engineering talent and long-term supplier commitments.
This is why the valuation cannot be judged only by a simple revenue multiple. Anthropic is trying to raise enough capital to compete in a race where infrastructure capacity is strategic. If the company can convert compute investment into better models, broader enterprise adoption and durable pricing power, a trillion-dollar valuation becomes easier to defend. If the industry’s spending arms race compresses margins or turns model access into a commodity, the same valuation becomes much harder to justify.
Timing, market conditions and the AI selloff
The IPO is arriving at an awkward but revealing moment. Axios reported that Anthropic shares are likely to begin trading sometime next week amid volatile AI trading conditions . CNBC, according to a summary of its November 10 report, said the company plans to price the IPO on November 18 and begin trading on Nasdaq on November 19, citing people familiar with the matter . AP said Anthropic is expected to make its Wall Street debut later this month .
That timing matters because the AI trade has become more contested. AP framed the offering against concerns about the heavy debt and capital flowing into the sector, while also noting that Anthropic, OpenAI and SpaceX’s AI business have all lost more money than they have made because of the high costs of building AI systems . Axios similarly noted that the company’s fully diluted valuation target comes after earlier speculation that it might seek as much as $2 trillion .
In that sense, the $1.3 trillion figure can be read two ways. It is an extraordinary target for a company founded in 2021 by former OpenAI leaders . But it may also represent a more disciplined number than the most aggressive expectations circulating earlier in the AI cycle. Public investors may reward that moderation if they see the IPO as priced to trade well rather than priced for maximum bragging rights.
Retail investors and liquidity risk
One notable reported feature is the potential retail allocation. CNBC’s report, as summarized by Readthrough, said Anthropic is targeting an unusually large 20% allocation for retail investors . If that holds, the IPO would not be marketed only to institutions; it would also invite broad participation from individual investors who have followed Claude and the AI boom as consumers, developers or market participants.
That could increase public enthusiasm, but it also raises the stakes. AI stocks can trade on narrative as much as near-term fundamentals, and a giant consumer-facing allocation could amplify early volatility. The filing’s lock-up language also deserves attention. Anthropic says directors, officers and holders of substantially all capital stock and convertible securities are subject to market standoff or lock-up agreements, with different periods applying to different holders and with some shares eligible for sale before the full lock-up periods expire .
For IPO buyers, the key question is not just the first-day pop or drop. It is the post-IPO supply schedule, the pace of future share unlocks, the company’s capital spending trajectory and whether revenue growth can keep outpacing dilution, compute costs and competition.
The central question for investors
Anthropic’s IPO is a referendum on the investability of frontier AI as a standalone public business. The company is seeking a valuation in the $1.3 trillion neighborhood, offering formal IPO terms of $70 to $78 per share and presenting itself as a high-growth but still unprofitable AI leader with a distinctive governance model .
The bullish case is straightforward: Claude has become one of the leading AI model families, enterprise demand is growing quickly, and the proceeds could give Anthropic the balance sheet to compete with the largest technology companies. The skeptical case is just as clear: the company is capital intensive, the sector may be overheated, governance rights are limited for ordinary shareholders, and public markets may be less forgiving than private investors.
That tension is exactly why this IPO matters. Anthropic is not only seeking a stock-market listing. It is asking public investors to decide what a frontier AI lab is worth before the economics of the industry have fully settled.