Anthropic, the creator of the Claude AI model, is preparing to sell shares to the public for the first time via an initial public offering (IPO), targeting a valuation of approximately $1.4 trillion, or £1 trillion. This would be the largest single capital raise in the corporate history of the United States. The company is selling approximately $65 billion worth of shares to the general public, alongside an additional $26 billion in shares designated for other major technology companies.
A valuation of $1.4 trillion would place Anthropic among the top 10 most valuable US companies. According to analyst estimates, the company has projected revenues of $52 billion in 2026 and $135 billion in 2027. Fiona Cincotta stated that the significance of this IPO lies in it being the first time public investors can access a pure AI investment, whereas previous offerings related to other parts of the ecosystem: semiconductors, chips and data centres.
However, Anthropic’s business carries risks. Approximately a quarter of revenue comes from a handful of companies, including Google and Amazon, while two-thirds of revenue is concentrated in the US. The company has contracts to purchase or lease hundreds of billions of dollars worth of chips and computing power from a few companies, including Broadcom, Google and Amazon. Many of these companies are simultaneously investors in Anthropic and major buyers of its products.
Gary Marcus, professor emeritus at New York University and author of the book Taming Silicon Valley, stated that AI is in a financial bubble and that the economics of AI do not hold water. According to him, the more companies build AI models, the more commoditised the technology becomes, which eliminates significant competitive advantages and lowers prices.
The background of this IPO dates back to November 2020, when Daniela Amodei called a friend in San Francisco seeking advice. Daniela Amodei had overseen safety and policy at OpenAI, where she and her brother Dario worked. The two had doubts about the safety levels at OpenAI, and the friend replied: "You don't need to be on the phone with me. You already know the answer." Amodei and several other colleagues resigned from OpenAI and jointly founded the rival company Anthropic.
Anthropic’s IPO comes at a time of market instability. Shares of companies supplying raw computing power for AI, such as Coreweave and Nebius, have fallen by half in value. Nvidia lost its title as the world’s most valuable company, and its shares fell 30 per cent, representing a drop in value of $2 trillion. Dario Amodei told Jamie Dimon, chief executive of JPMorgan Chase, at an event last week that his guiding principle is "never let your company be worth a single dollar more than it's worth".
Jensen Huang, chief executive of chipmaker Nvidia, resigned last week. He admitted to being "unfaithful" to his wife following reports that he had an affair with suspected Chinese spy Alice Huang. The fall of Jensen Huang and Nvidia’s 30 per cent drop in value from its peak have left AI investors searching for a new leader in the AI boom.
Growing public opposition to AI has become a factor in recent US midterm elections. Many areas in the US have imposed moratoriums on data centre development. Many candidates who were assessed as too close to major technology companies considered this a political burden rather than an advantage, while many Democratic candidates publicly opposed data centre development and won.













