Artificial intelligence developer Anthropic publicly filed for an initial public offering on Monday, seeking to list its shares on the Nasdaq under the proposed ticker ANT, after reporting a substantial increase in revenue alongside significant losses and long-term computing infrastructure commitments.
The company has appointed Goldman Sachs and Morgan Stanley to lead the offering but has not disclosed the number of shares it intends to sell or its proposed valuation.
According to The Wall Street Journal, citing people familiar with the plans, Anthropic could provide further offering details within approximately two weeks, following discussions with prospective investors.
The company is reportedly targeting a listing during the week of 16 November, although the timetable remains subject to change.
The newspaper reported that the offering could raise at least $100 billion and value Anthropic between $1.5 trillion and $2 trillion. These figures remain reported estimates rather than confirmed terms of the transaction.
Revenue Increases as Anthropic Reports $80.7 Billion Net Loss
Anthropic reported revenue of $16.3 billion for the six months ended 30 June, compared with $1.2 billion during the corresponding period a year earlier.
The increase reflects the expansion of the company’s artificial intelligence services, including its Claude models and enterprise products.
However, the filing also disclosed a net loss of $80.7 billion for the six-month period, compared with a loss of $13 billion a year earlier.
According to Reuters, citing the IPO prospectus, approximately $72 billion of the latest loss resulted from a non-cash accounting charge associated with changes in the valuation of convertible notes and other financing instruments.
The charge accounted for most of the reported net loss, although Anthropic also disclosed substantial operating and infrastructure-related financial requirements.
In its filing, the company cautioned that it may not achieve or maintain profitability, expects expenses to increase and anticipates further losses.
Anthropic also warned that its recent revenue growth rate may not continue and could decline as the business develops.
Enterprise Customer Base Expands
Anthropic reported a substantial increase in the number of business customers generating significant annual revenue.
The company had more than 11,900 customers contributing at least $100,000 in revenue, compared with approximately 950 a year earlier.
As of 30 September, more than 1,900 customers generated at least $1 million in revenue over the preceding 12 months, while approximately 250 customers contributed more than $10 million each.
The figures indicate an expansion in the number of businesses making substantial payments for Anthropic’s AI services.
Net revenue retention reached approximately 900% as of 30 September.
This measure reflects changes in revenue generated from existing customers, including increased spending, reduced usage and customer departures.
The reported figure indicates that revenue from Anthropic’s existing customer base had increased substantially over the measurement period.
However, the company cautioned that its historical growth may not be representative of future performance.
Anthropic Discloses $245 Billion in Future Payment Obligations
The IPO filing provides details of Anthropic’s substantial financial commitments associated with computing infrastructure.
The company disclosed approximately $245 billion in future minimum payment obligations, together with additional short-term agreements carrying around $21 billion in minimum payments.
Anthropic also outlined several arrangements with major technology and infrastructure providers.
Under its agreements with Google, the company expects spending of up to approximately $114.5 billion through 2033.
Its commitments involving Amazon total at least $110 billion through 2036.
Anthropic separately disclosed plans for up to $120.5 billion in computing expenditure involving SpaceXAI and Nvidia.
The company has also entered into five-year master equipment leasing arrangements with Broadcom, with total rental payments over the lease periods estimated at approximately $161.2 billion.
These amounts relate to different contractual arrangements and categories of obligations. They should not be added together to calculate a single total for future infrastructure expenditure because the disclosures may overlap.
The scale of the commitments reflects the computing capacity required to develop and operate large artificial intelligence models.
Anthropic expects capital expenditure to increase significantly and warned that its long-term cash requirements will remain substantial.
The company also noted that net cash consumption could rise even if its operating performance improves.
Cash Holdings, Credit Facilities and Additional Financing
Anthropic reported cash and cash equivalents of $112.4 billion as of 30 June.
The company had raised approximately $141.1 billion in capital by that date, according to its filing.
In September, Anthropic entered into a $15 billion revolving credit agreement, increasing its available revolving commitments from $3.3 billion.
The agreement also raised the company’s minimum liquidity covenant from $500 million to $5 billion and extended the facility’s maturity to September 2031.
Anthropic disclosed that it may issue up to $42 billion in convertible promissory notes to finance rental obligations.
Separately, Amazon agreed in April to provide up to $20 billion in financing under an arrangement described in the prospectus.
These financing measures provide additional resources as Anthropic expands its computing infrastructure and develops its AI products.
The company nevertheless warned that future capital requirements could remain considerable.
Corporate Governance and Public Benefit Structure
Anthropic is incorporated as a public benefit corporation, a legal structure that allows the company to pursue specified public benefits alongside its commercial objectives.
It has also established a Long-Term Benefit Trust as part of its governance arrangements.
According to the filing, Anthropic’s founders have pledged to donate 80% of their equity holdings to charitable causes.
The proposed governance structure includes a special Class T share category associated with Class T directors.
Anthropic also stated that it intends to keep its Claude AI service free from advertising, distinguishing its planned commercial approach from models that rely on advertising revenue from consumer usage.
The company does not expect to distribute dividends in the foreseeable future.
Anthropic Outlines Advanced AI Risks
Beyond its financial disclosures, Anthropic’s prospectus addresses risks associated with increasingly capable artificial intelligence systems.
The company identified the possibility of severe or potentially existential consequences from advanced AI development.
It also warned that some risks could increase as AI systems become more capable of automating research and development involving artificial intelligence itself.
The filing discusses potential scenarios in which advanced models could exhibit behaviour aimed at preserving their own operation, resist shutdown or manipulate information.
These scenarios are presented as potential risks rather than claims that such behaviour is common or has occurred in Anthropic’s deployed systems.
The disclosures form part of the company’s broader discussion of operational, technological and regulatory uncertainties.
Additional Investment and Share Transactions
Anthropic disclosed plans to invest $325 million in a newly established entity intended to accelerate the adoption of artificial intelligence among midsized businesses.
The company also reported that Amazon had purchased 19.3 million Series G-2 non-voting shares.
Separately, Anthropic repurchased approximately 4.5 million Series B shares from the United States in October 2025, according to the filing.
These transactions form part of the company’s disclosed financing and ownership arrangements ahead of the proposed public listing.
Potential $2 Trillion Valuation Draws Attention to AI Infrastructure Economics
If completed, Anthropic’s IPO would provide public-market investors with direct exposure to a major developer of large language models.
The company’s revenue growth, expansion of enterprise relationships and reported net revenue retention demonstrate the scale of recent customer spending on its services.
At the same time, its financial disclosures show substantial contractual obligations, expected increases in capital expenditure and uncertainty about future profitability.
The reported valuation range of $1.5 trillion to $2 trillion would place Anthropic among the world’s most highly valued publicly traded companies if achieved.
However, the final offering price, number of shares and valuation have not been announced.
Investors considering the proposed IPO will be able to assess the company’s revenue growth alongside its financing requirements, long-term computing commitments and expectations for continued losses.