Anthropic, an American company that developed the Claude AI and is one of OpenAI’s main competitors, officially filed documents on 26 October to list on the Nasdaq stock exchange. Its financial statements revealed rapid business growth: in the first half of 2026 alone, the company generated $16.29 billion in revenue. At the same time, the net loss for this period reached $80.7 billion, mainly due to an accounting revaluation of financial liabilities. According to Reuters, the forthcoming IPO will be a key test of investors’ willingness to fund the development of artificial intelligence, which requires hundreds of billions of dollars.
- On 26 October, Anthropic submitted an official application for an IPO in the US.
- In the first half of 2026, the company’s revenue amounted to $16.29 billion.
- The net loss amounted to $80.7 billion, but a significant proportion of this is attributable to non-cash accounting charges.
- The company plans to list its shares on Nasdaq under the ticker symbol ANT.
Anthropic has disclosed its financial results ahead of its stock market flotation
Anthropic, founded in 2021 by former OpenAI researchers, has become one of the leading developers of generative artificial intelligence. Its flagship product is the Claude family of models, which are used for programming, automating tasks and performing complex tasks.
The bulk of the company’s revenue comes from corporate clients, who pay for the use of the models based on the volume of data processed. Anthropic generates additional revenue from paid subscriptions to its services.
The financial figures point to exceptionally rapid business growth:
| Period | Anthropic’s revenue |
|---|---|
| 2023 | $41 million |
| 2024 | $386 million |
| 2025 | $4.59 billion |
| The first half of 2026 | $ 16.29 billion |
Consequently, in the first six months of 2026, Anthropic generated more than three times as much revenue as it did in the whole of the previous year.
However, as revenue increases, so do the costs of developing and operating artificial intelligence models.
Why did Anthropic report a loss of $80.7 billion?
According to the published financial documents, chAnthropic’s net loss for the first half of 2026 amounted to $80.7 billion. However, this figure does not mean that the company actually spent that much money on its operations.
Approximately $72 billion of the loss relates to the accounting revaluation of convertible financial instruments. Their estimated value increased in line with the rise in the valuation of Anthropic’s business itself.
The operating loss, which more directly reflects the results of core operations, amounted to $693 million for the first six months of 2026.
By way of comparison, for the whole of 2025, the company recorded an operating loss of $8.06 billion, whilst in 2024 the figure was $2.98 billion.
At the same time, even with high revenue growth, Anthropic remains dependent on significant investment in computing infrastructure.
The company has committed to more than $500 billion
The largest item of expenditure Anthropic — the computing power required to train and run Claude models.
The company rents server infrastructure, uses the Amazon Web Services and Google Cloud platforms, and invests in equipment for data centres.
According to Reuters, Anthropic has signed agreements with six partners involving expenditure of over $500 billion over a period of approximately ten years.
Among the largest liabilities are:
-
Amazon — at least $110 billion;
-
Google — approximately $114.5 billion;
-
Microsoft — $44.9 billion.
Some of these contracts provide for mandatory payments regardless of the actual use of the contracted capacity.
At the same time, Amazon and Google. They are not only Anthropic’s technology suppliers, but also major investors in the company. Amazon has already invested at least $13 billion and has agreed to invest up to $20 billion more. Google has committed to investing up to $40 billion.
This structure of collaboration attracts investors’ attention, as the money invested in Anthropic by technology corporations is partly returned to them through payments for cloud services.
Anthropic’s founders will retain control following the IPO
Despite the planned sale of shares, Anthropic’s management intends to retain decisive influence over the company’s governance.
A special class F share scheme will grant the seven co-founders a combined total of 50.1% votes on key corporate matters. Decisions on how these votes are to be used will be taken by a majority of the founders.
The company is headed by Dario Amodei, whilst his sister, Daniela Amodei, serves as president and chair of the board of directors.
Anthropic will also retain its status as a Public Benefit Corporation. This legal model requires the company to take into account not only the financial interests of its shareholders, but also its own social mission relating to the safe development of artificial intelligence.
In its investor documents, Anthropic specifically warned of the risks associated with the development of powerful AI systems, in particular their potential ability to manipulate information or resist being switched off in experimental conditions.
When will Anthropic’s shares be listed on the stock exchange?
The company plans to list its Class A shares on a US stock exchange Nasdaq under the ticker symbol ANT. The placement is being managed by the investment banks Goldman Sachs and Morgan Stanley.
However, the final number of shares, the price range and the amount to be raised have not yet been announced.
Some investors value Anthropic at approximately $1.5 trillion, although valuations of around $2 trillion had previously been discussed. The company’s final market value will depend on the terms of the placement and demand for its shares.
The forthcoming IPO will be a key test for the generative AI industry. Investors will have to assess whether the rapid growth in demand for intelligent systems can generate sufficient profit to cover the enormous long-term computing costs.
