Anthropic seeks Nasdaq IPO after $16.3 bn half-year revenue

Anthropic filed on Monday for a U.S. initial public offering on Nasdaq under ANT, disclosing $16.3 billion of revenue for the six months ended June 30 in a Form S-1 registration statement.

The filing gives public-market investors a large AI listing to assess: first-half 2026 revenue was more than three times full-year 2025 sales of $4.6 billion, while the half-year net loss reached $80.75 billion.

According to the filing, revenue was $41 million in 2023, $386 million in 2024 and $4.6 billion in 2025 before the first-half 2026 figure, which Anthropic said represented 1,300% year-on-year growth.

“We do not expect our revenue will continue to grow at comparable rates, and we cannot assure you that it will continue to grow at all,” the company cautioned in the filing.

Anthropic reported a $41.97 billion net loss for 2025, compared with the $80.75 billion loss for the first six months of 2026. The filing did not set the number of Class A shares to be sold or the proposed price range.

Goldman Sachs and Morgan Stanley are leading the underwriting group, which also includes J.P. Morgan, Citigroup, Barclays and Wells Fargo Securities. The proposed listing would put one of the best-known generative AI companies before public investors after a period of rapid sales growth and heavy losses.

Founded in 2021, Anthropic is a public benefit corporation, a legal form that permits directors to consider stated public-interest goals alongside shareholder returns. Its Long-Term Benefit Trust, described in the filing as “a financially disinterested trust,” generally can elect and remove a majority of the board.

Chief Executive Dario Amodei and President Daniela Amodei, together with five other co-founders, will hold special Class F stock through a founders’ entity. That structure gives the company a governance profile that differs from conventional single-class public companies.

Anthropic’s Claude chatbot is available through Amazon Web Services, Google Cloud and Microsoft Azure. The company said it trains models across Amazon Trainium, Google TPUs, Nvidia chips and AMD chips, tying its operations to several of the largest cloud and semiconductor suppliers.

The risk factors cited in the filing include immigration policy, chip export controls and concentrated advanced chip manufacturing in Taiwan. Anthropic said restrictions on professional and technical talent could hurt recruiting and retention, and warned that future U.S. export controls could raise costs, limit data-center locations or constrain overseas expansion.

The prospectus contains no specific disclosure on India-based operations or Indian customers. It says Anthropic has subsidiaries and operations in the U.K., the EU, Canada and Asia-Pacific, and that the prospectus has not been reviewed or approved by Indian regulators for a public offering in India.

The company has not determined the share count or price range for the IPO.