Anthropic Launches IPO Test for AI’s Trillion-Dollar Ambitions

Anthropic publicly filed for an IPO on Monday that could raise as much as $100 billion, potentially making it the largest stock-market debut ever.

The filing brings one of the largest private AI companies into public view while it remains unprofitable and spends heavily to expand its computing capacity. Its reception could affect valuations across the sector and the timing of rival OpenAI’s planned offering next year.

Anthropic submitted its papers to the U.S. Securities and Exchange Commission on 26 October 2026. It plans to list on Nasdaq under the ticker symbol "ANT".

The proposed fundraising would exceed the $86 billion raised by SpaceX’s IPO in June. The final amount is not known, so the headline figure is a ceiling rather than an agreed deal size.

Private investors valued Anthropic at $965 billion earlier this year. Reports have put a possible future valuation at $1.5 trillion or even $2 trillion, but no final valuation has been established.

Anthropic has not said how many shares it will sell or what price it will seek. Those missing terms mean the proposed raise and private-market valuation are not yet a public-market price that investors can assess.

The company remains unprofitable. Its loss for the first six months of 2026 was $80.7 billion, versus $13 billion in the same period a year earlier.

Revenue rose to $16.3 billion during the first six months of 2026, versus $1.2 billion during the previous year's first half.

Anthropic’s reported revenue was $41 million in 2023, $386 million in 2024 and $4.585 billion in 2025. It then recorded $16.286 billion in the first six months of 2026 alone.

The growth has required substantial investment. Anthropic spent $8.9 billion on chips and other infrastructure in the first half of the year, compared with $2.3 billion in the same period last year.

The company had $112.4 billion in cash and cash equivalents on 30 June. It also disclosed that it had raised $141.1 billion by the end of June and had incurred $6.2 billion in debt while building its infrastructure.

Anthropic said it had more than 14,000 enterprise customers as of 30 September. That included 11,900 customers generating $100,000 or more in revenue, more than 1,900 generating $1 million or more and about 250 generating $10 million or more.

Demand is not evenly distributed. A quarter of Anthropic’s revenue comes from its five largest customers, although no single customer accounts for more than 10% of its business.

The company has warned that it must use its existing AI models to design and train future platforms, a process it calls recursive self-improvement, or RSI, to remain competitive.

The cost of building AI has exceeded revenue at Anthropic and OpenAI, as well as SpaceX’s AI division. Investors have become more nervous about the sector’s financing needs, and several technology companies have recently withdrawn IPO plans.

Nigel Green, chief executive of investment firm deVere Group, said, "This IPO is a referendum on the entire AI trade". Anthropic’s debut would test whether public investors accept the sector’s unusually high spending requirements and whether enthusiasm extends to more listings.

Lightshed analysts said Anthropic faced an unusually broad range of possible outcomes: "Anthropic could become the most valuable company in the world. It could also end up in bankruptcy".

Sinan Aral, a professor at MIT’s Sloan School of Management, said the company would immediately rank among the world’s most valuable companies. He pointed to Anthropic’s rapid model development and its progress with enterprise customers.

Mark Daley, chief AI officer at Western University, said Anthropic’s economic importance would make pricing it difficult. "It’s really hard to value".

Anthropic’s filing also accelerates its contest with OpenAI. Former OpenAI leaders founded Anthropic in 2021, and OpenAI plans to go public next year while Anthropic is moving this year.

Anthropic filed confidentially with the SEC in June, allowing the agency to review its finances before the public submission. The company may list as early as November, but it has not confirmed that timetable.

The filing does not yet establish the final amount to be raised, the company’s final valuation, the offering price or the number of shares to be sold. It also does not make clear how much of Anthropic the founders’ nearly 500 million shares will represent after the IPO.

The seven co-founders, including Dario and Daniela Amodei, are seeking capital while retaining strong voting control. A special class of shares gives the founders 50.1% of shareholder votes.

The founders also appointed independent members of a Long-Term Benefit Trust. The members are financially disinterested and have the power to appoint directors, a structure intended to prevent financial interests from outweighing public safety.

In a letter to potential investors, the seven co-founders wrote, "We believe AI is one of the most consequential technologies in human history". They also urged prospective shareholders to view the purchase as more than a financial transaction.

The filing’s risk section runs nearly 100 pages. It discusses malicious actors using AI to build chemical weapons, authoritarian rulers using it against citizens and models going rogue with catastrophic or existential consequences for humanity.

In September, Dario Amodei said the AI industry should slow its rapid development so safety measures could catch up, warning that otherwise AI could release a swarm of agents capable of taking over the entire internet within six to 12 months.

Earnest Wong, a portfolio manager and head of research at Baskin Wealth Management, said the prospectus gave less prominence to Amodei’s public safety warning than he expected. "Most of the risks are kind of just normal, vanilla legal disclosures that are included in most companies".

Anthropic said that, in September 2025, staff recorded what they consider the first large-scale cyber-attack without significant human intervention. A Chinese state-sponsored hacking group manipulated its Claude Code tool into attempting to infiltrate roughly 30 global targets.

Anthropic said it could use some IPO proceeds to acquire other businesses and technology and to invest in infrastructure as it competes with rivals.

The company is also in a dispute with the Trump administration over the use of its products in warfare. The dispute led to Anthropic being designated a supply chain risk, which bars military agencies from contracting with it. Anthropic has challenged the blacklisting.

The public offering will make Anthropic’s pledge to prioritise public safety answerable to shareholders, while its voting structure gives the founders and the Long-Term Benefit Trust unusual influence over the company’s direction.

1. JuneConfidential filing

   Anthropic filed confidentially with the SEC for review of its finances.

2. September 2025Cyber disclosure

   Staff documented what they believed was a large-scale cyber-attack involving Claude Code.

3. SeptemberSafety warning

   Dario Amodei urged the AI industry to slow development so safeguards could catch up.

4. 26 October 2026Public filing

   Anthropic filed publicly with the SEC and set out its proposed Nasdaq listing under ANT.

5. November 2026Possible IPO

   The company could list as early as November, although the timing remains unconfirmed.

The next market test is the release of final share, price and valuation details. A possible IPO as early as November 2026 is the next dated step, although the timing remains unconfirmed.