Anthropic conducted the largest IPO in history, raising $81.3 billion after increasing its offering size. This allowed the chatbot developer, Claude, to be valued at $1.5 trillion, despite concerns about the AI market and fundraising in the tech sector.
On Wednesday, Anthropic set its stock price at $83 per share, exceeding the initial range of $70–$78, indicating high investor demand. The company increased the number of shares offered by approximately 19%, bringing it to 980 million. Thus, the total deal size reached $107.3 billion, including an additional $26 billion that Anthropic is raising through a private placement at the IPO price, according to documents.
With the fully diluted share count, including options and restricted stock, the company's valuation is about $1.5 trillion, and on a non-diluted basis, it is about $1.1 trillion. Anthropic plans to begin trading on Nasdaq on Thursday under the ticker ANT.
This listing comes amid a complicated situation for AI stocks. In recent weeks, chip manufacturers and cloud companies that raised hundreds of billions of dollars to fund spending plans have faced pressure due to valuation concerns, rising AI infrastructure costs, and Jensen Huang's sudden departure from Nvidia.
Armina Rosenberg, portfolio manager and co-founder of Minotaur Capital, who is expecting Anthropic shares, noted: 'There is a real barrier of concern right now, and several things will be needed to overcome it. Once Anthropic publishes its report, it will be one of the clearest tests of whether the economic foundations of AI hold up.'
A successful launch could boost confidence in other companies awaiting public market debuts. Anthropic's nearest competitor, OpenAI, is expected to aim for a listing as early as next year.
Spending Commitments Reveal Risks
Anthropic's growth is driven by high demand for its Claude models, especially Claude Code, which is used as a coding agent and is popular among software developers. According to the IPO documents, the company's annual revenue reached $81 billion as of October 31, which is $10 billion more than on September 30.
As of September 30, the company served 11,900 customers generating at least $100,000 in annual revenue over the last 12 months, higher than the more than 1,500 at the end of last year.
However, maintaining this growth is becoming expensive. In the first half of 2026, Anthropic spent nearly $9 billion on computing power and infrastructure, almost four times the amount from the previous year. In the third quarter, the company incurred another $5 billion in computing costs, which reduced the revenue generated per dollar of computing power.
Third-quarter revenue grew by 47% to $16.96 billion compared to the second quarter. The company reported an operating loss of $567 million, compared to an operating profit of $974 million in the previous quarter. Net loss decreased to $23.3 billion from $66.3 billion in the quarter prior, including non-operating items related to convertible bonds.
Anthropic has committed to nearly $900 billion in computing expenses with infrastructure partners, including Amazon and Google from Alphabet. The company warned that these partners rely on third-party funding to acquire chips and build capacity. In the documents, Anthropic stated: 'If such funding or support becomes unavailable or more expensive, the agreed-upon capacity may be postponed, revalued, reduced, or not provided.'
Competition and Safety
Anthropic faces questions about how to balance its commercial ambitions with its focus on AI safety. CEO Dario Amodei called on the industry to slow down the pace of developing and releasing increasingly powerful models, warning that AI advancements may outpace society's ability to manage risks.
Competition is also intensifying. OpenAI's latest flagship model, GPT-6 Astra, accounted for about 13% of corporate AI spending tracked by the expense management platform Ramp, compared to about 8% for Anthropic's Claude Fable, according to data cited by Reuters in September. Reuters also reported that Meta Platforms, one of Anthropic's largest clients, is considering reducing its use of the startup's models as it develops more internal AI capabilities.
Peter Singlhurst, head of private companies at Baillie Gifford, who invested about $1.5 billion in Anthropic before its IPO, stated: 'Our hypothesis is that most of the value will go to those at the forefront.' He added: 'Our bet is that Anthropic has the ingredients—it doesn't guarantee it, but it means they might have the highest probability of staying at the forefront.'



