Anthropic IPO filing puts AI spending, valuation and profitability in focus
Anthropic has published its plan for an initial public offering (IPO), or the first time a company sells its shares to the public. This will test whether investors are still willing to put money into AI companies as concerns grow about the high cost of building advanced AI models.
According to its filing with the US Securities and Exchange Commission, the company behind Claude plans to list on the Nasdaq stock market under the ticker symbol “ANT”.
Goldman Sachs and Morgan Stanley are leading the group of banks managing the share sale.
The IPO comes as investors take a closer look at the AI boom, which has been supported by borrowing. They are questioning whether spending on data centres, chips and computing systems will bring in enough money to justify the high values placed on AI companies.
Revenue is growing, but the financial costs are high
Anthropic's prospectus shows how quickly the company has grown, thanks to demand for its AI models and coding tools, including Claude Code.
The company reported revenue of about $16.3 billion in the first half of 2026, while its net loss grew to $80.7 billion.
Most of this loss came from an increase in the estimated fair value of bonds that can be converted into company shares.
Anthropic reportedly told investors that its third-quarter revenue rose by nearly 50% from the previous quarter to $16.8 billion. The company also said it made a profit when measured using adjusted figures, which leave out certain costs.
The company's net loss grew to $80.7 billion in 2026, compared with a loss of $41.9 billion in 2025 and $8.3 billion in 2024.
Anthropic was founded in 2021 by former OpenAI researchers and engineers. It has since become one of the main competitors in the market for generative AI, which creates new content such as text and code.
Amazon and Google were among its early business partners. They provided billions of dollars in investment and access to cloud computing systems.
Anthropic said in its prospectus: “We believe this abundance of AI will rapidly transform every sector of the global economy in ways greater than the impact of industrialization, electricity, semiconductors and the internet.”
However, keeping up this growth requires huge investments in the computing systems needed to build and run increasingly advanced AI models.
Expectations for the company's value face a tougher market
Anthropic's IPO is expected to give investors a useful point of comparison when deciding how much leading AI companies are worth. These include its rival OpenAI, which filed confidentially for an IPO in June.
Anthropic was valued at $965 billion in a funding round in May.
However, Reuters reported that some existing and potential investors valued the company at about $1.5 trillion, down from around $2 trillion just one week earlier. The change followed the release of financial details at an event for investors.
These changing expectations come as stock markets around the world face pressure from concerns about AI spending, higher borrowing costs and rising bond yields, which affect the cost of borrowing and investment returns.
Companies that provide AI computing infrastructure have also seen sharp changes in their market values.
CoreWeave and Nebius each lost more than 30% of their value in October as concerns grew about the risk of unpaid debt across the data centre industry.
Anthropic's planned listing follows SpaceX's record-setting IPO in June, which raised $85.7 billion.
Selling shares to the public could help Anthropic raise more money to fund its growth.
However, the company's value will depend on how investors judge its growth prospects, ability to make profits and ability to remain competitive.
AI safety adds another risk
Along with financial risks, Anthropic's prospectus warns about dangers linked to increasingly advanced AI systems.
The company warned that its models could show “self-preserving behaviors”. This could include attempts to “resist shutdown”, “conceal or manipulate information”, or act in ways “resembling blackmail”.
These warnings come as AI developers face questions about how to control the risks of increasingly powerful systems while making them available for commercial use.
Chief Executive Dario Amodei has urged the global AI community to slow the release of new AI capabilities to address safety concerns.
Even so, Anthropic has continued building its AI models to compete with OpenAI and other rivals.
OpenAI Chief Executive Sam Altman said in September that the company would not go public in 2026, pointing to concerns about AI safety.
He said that even a 10% chance of AI causing human extinction by the end of the decade was “unacceptable”.
Anthropic's IPO will therefore test more than whether investors want to buy shares in a fast-growing AI company.
It will also give investors a chance to judge whether the company's growth, heavy need for computing infrastructure and approach to AI safety can support the high value expected of it on the public stock market.