Anthropic investors expect the company to go public in October at a valuation of $2 trillion or more, which would make it the largest initial public offering in history, according to the Financial Times. The figure would surpass SpaceX, which went public at $1.77 trillion in June 2026 and currently holds the record for the largest IPO ever completed. Six Anthropic backers told the Financial Times that the company's rapid revenue growth could support a valuation more than twice its most recent level, set at $965 billion in May after new investment. Morgan Stanley, Goldman Sachs, and JPMorgan are leading the offering.
The bull case runs through a single extraordinary growth curve. Annualized run rates have skyrocketed from $1 billion in late 2024 to $47 billion by May, with projections of $100 to $120 billion by December. "If Anthropic is growing 800 per cent a year, you'd think at the incredibly low end they would trade at 30 times revenue," one investor told the FT, a calculation that would value the company at $3 trillion. Investors have looked to companies benefiting from AI spending, including Palantir and cloud provider Nebius, which have traded at roughly 55 times revenue this year, as the nearest public comparables. The enterprise foundation underpinning those projections is concrete: the number of customers spending over $100,000 annually on Claude has grown sevenfold in the past year, and over 1,000 customers now spend more than $1 million annually, doubling from 500 in under two months as of April 2026. Anthropic also surpassed OpenAI in U.S. business AI spending for the first time in April 2026, capturing 34.4 percent of companies on the Ramp AI Index versus OpenAI's 32.3 percent.
The accounting caveat, however, is material. The $100 to $120 billion figure is based on Anthropic's preferred annualized run-rate measure, which extrapolates recent sales performance over a full year rather than representing audited revenue generated during a 12-month period. On actual calendar-2026 revenue, Anthropic is expected to land somewhere around $20 to $26 billion for the year. The projections come from investors rather than Anthropic itself; senior executives have not established an IPO valuation target even in private conversations, and investors have instead produced their own financial models based largely on the company's recent growth. Anthropic's confidential S-1, filed with the SEC on June 1, 2026, under the JOBS Act's emerging-growth-company provisions, means that a public prospectus, when released, will be the first time investors can scrutinize audited financials, a full risk-factor disclosure, and the company's voting governance structure.
The risks that prospectus will need to address are already visible. Anthropic's top model carries a price tag more than 2.5 times higher than OpenAI's flagship offering, while Chinese open-weight alternatives can be accessed at a fraction of that cost. Anthropic has clashed with the Trump administration on multiple occasions and is currently in litigation with the U.S. Department of Defense, which designated Anthropic a supply-chain risk entity; in June, the Department of Commerce imposed export controls that forced Anthropic to temporarily withdraw its top-performing Mythos model, a move investors say slowed overall revenue growth for that month. Compute costs add structural pressure: Anthropic's estimated $19 billion in compute spend in 2026 is approximately 40 percent of its $47 billion annualized run rate. OpenAI, which also filed confidentially around the same period, has shifted toward a potential 2027 window according to multiple reports, giving Anthropic a possible first-mover position among pure-play frontier AI labs, a strategic advantage that is real but independent of underlying fundamentals.
The single most important document in AI finance this year will not be a research paper; it will be Anthropic's public S-1, which will force a reckoning between the run-rate story investors are telling themselves and the audited numbers they will finally have to defend.