Anthropic’s $965 Billion IPO Carries a Catch for Buyers

Anthropic, the maker of the Claude chatbot, said on June 1 that it had confidentially filed draft paperwork for an initial public offering with U.S. securities regulators, the strongest signal yet that the most valuable AI startup wants public money. The filing lines Anthropic up for an IPO near a $965 billion valuation, alongside SpaceX and OpenAI in a sudden race to Wall Street.

On its face the move reads like pure validation for Anthropic and for AI investing broadly. The fine print is more demanding: public buyers would take a stake in a company that loses billions a year, does not expect to generate its own cash until near the end of the decade, and is structured so a safety-focused trust can outvote them on who controls the board.

A Confidential Filing Restarts the IPO Machine

Anthropic confirmed the step in a short statement, saying it had “confidentially submitted” a draft registration statement on Form S-1 (the document a company files before selling shares to the public) to the Securities and Exchange Commission (SEC, the U.S. markets regulator). A confidential filing lets a company start the review clock without showing rivals its books. “This gives us the option to go public after the SEC completes its review,” the company said, adding that share count and price had not been set.

The move turned a quiet IPO calendar into a sprint. Elon Musk’s SpaceX filed its own S-1 on May 20 and headed into its investor roadshow this week, and OpenAI, the maker of ChatGPT, is widely reported to be preparing a confidential filing of its own. Analysts at Wedbush Securities called it a turning point. “We believe this represents an opening of the floodgates for the IPO market, which has been relatively dormant for a few years,” wrote Dan Ives, a managing director at the firm.

Order matters in a race like this. Both frontier labs will be asking public markets for tens of billions of dollars within months of each other, and the company that prices first tends to drink first from a finite pool of capital. The filing came barely a week after Anthropic’s leap past OpenAI on paper, and getting to the regulator first keeps that lead intact.

A $965 Billion Price Tag on Billions in Losses

The valuation comes straight from private money. Anthropic’s $65 billion Series H round, which closed on May 28, was led by Altimeter Capital, Dragoneer, Greenoaks and Sequoia Capital. That round, not a public market, set the number a Wall Street listing would now have to defend.

• $965 billion post-money valuation from the Series H round, the highest ever for a private AI company.
• About 20 times the company’s roughly $47 billion annualized revenue run-rate, a steep multiple even by enterprise-software standards.
• Roughly $14 billion in projected losses for 2026, according to internal financial figures reported this spring.

The growth that justifies the optimism is genuine. Run-rate revenue climbed from about $9 billion at the end of 2025 to more than $47 billion this spring, a more than fivefold jump in months, and the company has told investors annualized revenue should clear $50 billion within weeks. Few businesses of any size scale that fast.

The cost side is where the story gets heavier. Anthropic has signaled infrastructure commitments running past $1 trillion over the next several years, and reported internal projections do not show positive free cash flow (FCF, the cash left after operating costs and capital spending) until around 2029. A buyer at the listing would be paying a near-trillion-dollar price for a company that plans to lose money for roughly three more years before the math turns.

That gap is the contrarian heart of the deal. The revenue curve is real, the customer demand is real, and so is the multi-year stretch of red ink the prospectus will have to explain. Public investors price both at once.

The Control Catch Buried in the Structure

The floodgates story tends to skip the structure. Anthropic is a Delaware public benefit corporation (PBC, a for-profit whose charter lets directors weigh a public mission against shareholder returns), and that charter changes what a share actually grants.

What Class T Stock Controls

Sitting above ordinary shareholders is the Long-Term Benefit Trust (LTBT), an independent body of five financially disinterested members chosen for backgrounds in AI safety, national security and public policy. The trust holds a special class of stock that hands it growing power over board seats.

The Class T stock grants the Trust the authority to elect and remove a number of Anthropic’s board members that will phase in according to time- and funding-based milestones.

That is Anthropic describing its own Long-Term Benefit Trust design. By the company’s account, the trust will elect a majority of the board within four years, a level of insulation from ordinary stockholders that almost no public company carries.

A Mission That Can Outrank Earnings

The practical effect is blunt. As a public benefit corporation, Anthropic’s directors are allowed to balance its safety mission against profit, which means shareholders cannot force the company to drop safety priorities just to lift a quarter’s earnings. The trust cannot fire the chief executive or any employee, but it shapes who sits on the board that does.

For a retail buyer used to one share, one vote, this is unusual territory. You can own the upside of Claude and still hold limited say over the people steering the company, an arrangement most index investors will absorb without ever reading the governance section.

Three Trillion-Dollar Names Racing the Same Window

The contest is really three companies chasing the same investors through the same narrow opening, each with a different business and a different reason to move now.

Anthropic’s valuation eclipsed its closest rival on paper for the first time in late May, and with the filing now lodged, the sequencing matters as much as the numbers. Whoever prices first taps public capital before the others, which is why a confidential filing landed before a single share count was set.

Why Demand Is Real Even If Profit Isn’t

None of the caution erases the demand pulling investors in. The customer base behind Claude is large, paying and growing, which is exactly what makes the losses tolerable to the funds writing checks.

• Run-rate revenue that jumped from about $9 billion at the end of 2025 to more than $47 billion this spring.
• Salesforce’s plan to spend around $300 million on Anthropic’s models in 2026, even while freezing engineering hires.
• A Series H round led by Altimeter Capital, Dragoneer, Greenoaks and Sequoia Capital, with sovereign and crossover funds crowding into the co-lead group.
• Enterprise adoption the company expects to push annualized run-rate past $50 billion within weeks.

The enterprise pull is concrete, not hypothetical. Salesforce’s $300 million commitment to Anthropic is the kind of contract that turns a model into a revenue line, and it lands at a moment when big software firms are routing budgets toward AI tools instead of new headcount.

Anthropic frames the cash it keeps raising as fuel for that demand. “This funding will help us serve the historic demand we are experiencing, stay at the research frontier, and bring Claude to more of the places where work happens,” said Krishna Rao, the company’s chief financial officer. The bet investors are making is that the demand keeps compounding faster than the compute bill.

What the SEC Clock Means for a Fall Debut

A confidential filing is a start, not a finish. The regulator now reviews the draft privately, often through several rounds of comments, before the company can flip the prospectus public and set a price. That is why Anthropic stressed that any offering “will depend on market conditions and other factors.”

On the standard timeline, a confidential submission this spring could clear review and reach the market by the fall, which is roughly the window analysts are penciling in for the trio. The SEC’s going-public process gives a company room to wait for a friendly tape, so the schedule is a target, not a promise.

For buyers, the decision underneath the hype is simpler than the headlines suggest. You are paying a near-trillion-dollar price for surging revenue, a years-long stretch of losses, and a board that a safety trust will largely control. Americans are split on AI’s role in daily life even as its commercial machine accelerates, and a public listing puts that argument on a ticker.

If demand and markets hold through the regulator’s review, Anthropic could be trading by the fall and the floodgates Ives described open behind it. If the appetite that set the valuation cools before the prospectus goes public, the same filing that started the race becomes the reason it stalls.

Disclaimer: This article is for informational purposes only and is not investment advice. Securities and pre-IPO companies carry significant risk, including the loss of capital, and valuations cited here reflect private funding rounds that may not match any future public offering price. Consult a qualified financial professional before making investment decisions. Figures are accurate as of publication.