What happened
Anthropic set the final price range for its initial public offering at $70 to $78 per share, targeting a valuation of $916 billion to $1.02 trillion on existing shares and up to $1.4 trillion when including all potential future issuances. The public offering could raise as much as $64.4 billion, and together with a $26 billion private placement from investors including Nvidia, Google and Microsoft, the total capital raise may reach $90.4 billion—surpassing SpaceX’s record IPO earlier in the year. The company disclosed rapid revenue growth, reporting $17 billion in Q3 2026 (a 1,127 % year‑over‑year increase) and annualized revenue estimates of $47 billion in May and $81 billion in October. At the same time, Revolut, a UK‑based digital bank with about 2,000 engineers, replaced Anthropic’s Claude Code with Zhipu AI’s GLM 5.3 model, citing a ten‑fold cost reduction. Revolut runs GLM on its own U.S. servers and claims to spend only a tenth of what it paid for Claude Code and OpenAI’s offerings.
Anthropic announced that it will price 825 million shares at $70‑$78 each, setting a valuation range of $916 billion to $1.02 trillion on the shares currently outstanding. Including all shares that could be issued under the prospectus, the upper bound reaches $1.4 trillion.
The public offering could raise up to $64.4 billion. Combined with a $26 billion private placement led by Nvidia, Google, Microsoft and four other investors, the total capital raised may total $90.4 billion, eclipsing SpaceX’s $85.7 billion IPO earlier in 2026.
Revenue growth is highlighted in the prospectus: Q3 2026 revenue was $17 billion, a 1,127 % increase from the prior year. Annualized revenue estimates rose from $9 billion at the end of 2025 to $47 billion in May 2026 and $81 billion in October 2026.
Operating expenses surged, with Q3 and infrastructure costs climbing 77 % to $9.9 billion, half of which was sourced at premium spot prices. The company posted an operating loss of $567 million for the quarter, and price cuts reduced revenue by roughly $600 million between Q2 and Q3.
Revolut, a UK digital bank with roughly 2,000 engineers, switched its standard AI coding assistant from Anthropic’s Claude Code to Zhipu AI’s GLM 5.3 model, citing a ten‑fold reduction in cost. Revolut runs GLM on its own U.S. servers using 56 Nvidia B200 chips and claims to spend only a tenth of what it previously paid to Anthropic and OpenAI.
Why it matters
The Anthropic IPO is a litmus test for the broader $1.4 trillion AI spending bet made by investors, governments and chip makers. If the market accepts the high valuation, it could validate the business model of large‑scale foundation‑model providers that rely on massive spending and long‑term contracts with cloud and chip partners. Conversely, the shift by a high‑profile client to a cheaper, open Chinese model highlights price sensitivity and the risk that cost‑effective alternatives could erode margins for U.S. firms. The IPO also ties together several market dynamics: Nvidia’s $10 billion private placement with Anthropic, the potential impact of U.S. export controls on model availability, and the broader question of whether AI companies can sustain valuations that imply revenue multiples far above historic IPO norms. Analysts note that only 14 of more than 4,000 U.S. IPOs since 1980 have debuted at comparable price‑to‑sales ratios, and those have underperformed the market by an average of 15.4 % over three years. The outcome will influence capital allocation decisions across the AI ecosystem, from venture funding to corporate AI budgeting.
The IPO’s valuation tests whether investors are willing to bet on AI firms that require billions in annual spend and still operate at a loss. A successful debut could unlock further capital for AI research, chip manufacturing, and cloud infrastructure, reinforcing the $1.4 trillion AI spending narrative that underpins recent private‑equity and sovereign‑wealth fund commitments.
The price‑to‑sales multiple implied by the upper valuation exceeds historic IPO norms. Only 14 of over 4,000 U.S. IPOs since 1980 have debuted at similar multiples, and those have underperformed the broader market by an average of 15.4 % over three years, suggesting heightened risk for investors.
Revolut’s migration to a cheaper Chinese model underscores growing price pressure on U.S. AI providers. If more enterprises adopt open‑, lower‑cost alternatives, Anthropic’s high‑margin assumptions could be challenged, potentially compressing future revenue growth and profitability.
Nvidia’s $10 billion private placement ties the chipmaker’s fortunes to Anthropic’s success. A weak IPO could diminish confidence in Nvidia’s AI‑related revenue outlook, while a strong debut may buoy Nvidia’s stock and support its broader AI ecosystem strategy.
Regulatory dynamics, such as the June 2026 U.S. export controls that temporarily halted Anthropic’s model access, remain a wildcard. Continued scrutiny could affect the company’s ability to serve foreign customers and impact future revenue streams.
Interactive Mechanism: How It Actually Works
Explore the underlying technology behind this development interactively.
Which component of an AI application is the machine-learning model itself?
What to watch next
Investors will monitor the final pricing on November 18 and the first day of trading on November 19 to gauge market appetite for ultra‑high‑valuation AI firms. Key metrics to watch include post‑IPO share performance, the pace of institutional demand (the offering was reportedly six‑times oversubscribed), and any shifts in Nvidia’s stock as its private placement terms become public. Analysts will also track whether other AI‑focused companies, such as OpenAI, Databricks or Stripe, adjust their IPO timelines in response to Anthropic’s market reception. Finally, the adoption of lower‑cost Chinese models by enterprises like Revolut could spur further cost‑competition, prompting U.S. providers to adjust pricing or expand open‑ offerings.
Final pricing on November 18 and first‑day trading performance on November 19, especially whether the shares close above the $78 upper range.
Institutional demand levels, given reports that the offering was six‑times oversubscribed, and any adjustments to the share allocation between retail and institutional investors.
Nvidia’s Q3 fiscal earnings (released November 17) for clues on how the private placement with Anthropic is reflected in its revenue guidance and stock price.
Potential ripple effects on other AI‑centric IPO plans, notably OpenAI’s confidential filing, Databricks’ 2027 target, and Stripe’s delayed listing, as they gauge market appetite.
Further cost‑competition signals, such as additional enterprise migrations to Chinese open models (e.g., DeepSeek, Qwen) and any pricing adjustments announced by Anthropic or OpenAI in response.