Anthropic's IPO Story Changed With the Amended S-1

Anthropic's public listing is no longer simply a story about one of the world's fastest-growing AI companies coming to market. The amended S-1 filed November 2 gives investors a much clearer look at the economics behind Claude, and the numbers are striking in both directions. Revenue reached $16.96 billion in the third quarter, up from $11.56 billion in the second quarter, while nine-month revenue reached $33.24 billion. At the same time, the company disclosed approximately $895.4 billion of aggregate purchase and lease commitments, substantially all tied to compute and infrastructure.

The revenue growth is extraordinary, but the customer figures show that it is not confined to a small group of AI enthusiasts. More than 11,900 customers generated at least $100,000 of revenue on a trailing 12-month basis at September 30, compared with more than 950 a year earlier. More than 1,900 customers generated at least $1 million and about 250 generated at least $10 million. Anthropic also reported net revenue retention of approximately 900%, indicating that existing customers are materially expanding their usage.

The first valuation question is the rate of deceleration. Revenue increased 47% from Q2, according to the amended filing, after growing 144% in Q2. A 47% sequential increase is still exceptional, but the slowdown matters when a company is approaching an IPO at a potential valuation of $1.5 trillion or more. Public investors will be paying for years of future growth, not merely the revenue already on the books.

Quarterly revenue. Source: DaveFinances.com.

The Growth Is Real, but the Cost Curve Is Moving Too

Anthropic's operating figures explain why this IPO will be judged differently from a conventional software listing. The company reported operating income of $974 million in Q2 but swung to a $567 million operating loss in Q3. Compute and infrastructure expense reached $9.86 billion in Q3, versus $5.56 billion in Q2 and $3.39 billion in Q1.

The net-loss figure is even more dramatic, although it needs to be read carefully. Anthropic reported a $23.29 billion net loss in Q3, but the filing says changes in the fair value of convertible notes, derivative liabilities and SAFEs produced a $94.3 billion non-cash charge for the first nine months of 2026. That accounting adjustment is largely connected to the rising value of the company's financing instruments. It is not the same thing as Anthropic burning $104 billion of cash.

That distinction does not make the underlying economics irrelevant. The operating line is where investors should focus. Anthropic spent heavily enough on compute and infrastructure to move from operating profitability in Q2 to an operating loss in Q3 even as revenue climbed by more than $5 billion sequentially. The company is proving that customers will pay for Claude, but it is also showing that supplying the capacity required to serve them remains extraordinarily expensive.

Figure 2. Reported operating income/loss. Source: DaveFinances.com.

The $895 Billion Commitment Is the Number Investors Cannot Ignore

The headline figure in the amended filing is approximately $895.4 billion of aggregate committed spend. Anthropic says substantially all of it relates to compute and infrastructure. Roughly $447.5 billion is expected to be incurred before the end of 2030, with substantially all of the remainder falling between 2031 and 2045. About $69.8 billion is due in the 12 months following September 30, 2026.

That does not mean Anthropic has borrowed $895 billion, or that the company will write a check for that amount tomorrow. The commitments include purchase agreements, leases that have not yet commenced and arrangements with termination rights or conditions. The filing says approximately $213.8 billion of the aggregate amount is subject to exercisable termination rights or contingent conditions, including counterparties securing financing, facilities or regulatory approvals.

The qualification matters, but so does the reason Anthropic is making these commitments. Frontier AI is becoming an infrastructure business as much as a software business. Training and inference require huge quantities of accelerators, data-center capacity, networking and electricity, so Anthropic has to secure capacity before the revenue arrives. The result is a financial structure in which future growth is increasingly tied to long-duration infrastructure commitments.

Anthropic also warns that some compute providers depend on third-party financing or credit support to acquire chips and develop capacity. If that financing becomes unavailable or more expensive, contracted capacity could be delayed, repriced, reduced or not delivered. That puts part of Anthropic's supply chain in the hands of capital markets it does not control.

Aggregate compute and infrastructure commitments by disclosed timing bucket. Source: DaveFinances.com.

Amazon and Google Are More Than Passive Investors

Anthropic's IPO gives public-market investors a new way to think about Amazon and Alphabet. Both companies have backed Anthropic financially, but the relationship goes beyond a venture investment because cloud infrastructure is central to Anthropic's business. Amazon has invested about $18 billion through convertible notes and preferred equity, according to reporting based on company disclosures, while Amazon's notes could convert into as many as 1.84 billion shares subject to a 12% ownership cap unless Amazon waives it.

The strategic connection matters because Anthropic's infrastructure commitments include major relationships with cloud and technology providers. Amazon's AWS is a key infrastructure partner, while Google is another important provider and investor. That creates an unusual loop: Anthropic needs capital and compute to grow; its backers can benefit from the value created by that growth; and those same companies can generate revenue by supplying the infrastructure required to run the models.

For investors, the question is whether that structure creates an efficient AI ecosystem or simply moves money around an extremely expensive supply chain. Anthropic's success can benefit Amazon and Alphabet through equity appreciation, cloud demand and strategic positioning. If AI spending becomes less profitable, however, the exposure runs in both directions.

Enterprise Adoption Is the Strongest Bull Case

The strongest evidence supporting Anthropic's valuation is not the size of the AI market. It is the depth of customer expansion disclosed in the filing. The number of customers generating at least $100,000 on a trailing 12-month basis increased from more than 950 in September 2025 to more than 11,900 a year later. Revenue from that customer group reached $24.5 billion for the nine months ended September 30, representing about 74% of total revenue.

Anthropic's approximately 900% net revenue retention is another important figure. Existing customers are not merely renewing; many are expanding their Claude deployments across departments and workflows. Products such as Claude Code and other agentic applications could push that expansion further if companies move from experimenting with AI assistants to assigning models production work.

The catch is concentration and cost. A large share of revenue comes from customers spending at least $100,000 a year, while compute costs rise as usage expands. High retention is valuable only if the incremental revenue carries enough margin to cover the infrastructure required to deliver it. That is the economic test Anthropic now has to pass in the public market.

Customers generating at least $100,000 of trailing-12-month revenue. Source: DaveFinances.com.

The IPO Will Be Priced on Future Margins, Not Current Growth

Anthropic has not disclosed the final number of shares or the price range, so investors cannot calculate a definitive IPO valuation from the S-1 alone. Reports have put potential valuations around $1.5 trillion to $2 trillion, but those figures remain market expectations rather than an announced offer price. The eventual pricing will tell investors how much of the growth story has already been capitalized into the stock.

At a $1.5 trillion valuation, investors would be paying roughly 45 times nine-month revenue annualized using the $33.24 billion figure as a simple run-rate reference. At $2 trillion, the figure rises to roughly 60 times. Those are not conventional mature-software multiples. They only make sense if Anthropic can sustain extraordinary revenue growth while improving the economics of each dollar of compute.

That is why the Q3 slowdown deserves more attention than the headline revenue jump. A 47% sequential increase is still enormous, but the company is simultaneously committing hundreds of billions of dollars to infrastructure and warning about financing risks among suppliers. The market has to decide whether that is the necessary price of building a frontier AI company or evidence that revenue growth is being purchased at an unsustainably high cost.

What to Watch When Anthropic Starts Trading

The first datapoint will be the IPO price relative to the valuation investors are willing to accept. The second will be how the stock trades after the initial allocation: a huge first-day jump could signal scarcity and enthusiasm, but it would not prove the underlying economics are sound. The more useful test will come with subsequent quarterly reports, when investors can compare revenue growth against compute spending, operating margins and the utilization of the infrastructure Anthropic has committed to.

Anthropic is entering the public market with genuine enterprise traction, a rapidly expanding revenue base and one of the strongest positions in the frontier-AI race. It is also arriving with an infrastructure bill that is almost impossible to ignore. The S-1 does not show a company running out of demand; it shows a company trying to scale fast enough to meet demand while making enormous bets on future compute availability.

That makes the IPO a much bigger test than whether investors like Claude. The real question is whether frontier AI can turn extraordinary demand into durable, high-margin cash flow before the infrastructure commitments become the dominant financial fact of the business.