Anthropic's S-1 Shows $16.3 Billion Revenue and an $80.7 Billion Loss

Anthropic reported $16.3 billion in revenue for the first half of 2026 and an $80.7 billion net loss in a public filing that opens the way to a Nasdaq listing under the ticker ANT. The company filed its Form S-1 on Oct. 26, and six days later the document is still the main source of hard numbers. Next steps could arrive within weeks, which makes this the AI market story to track now.

The filing does not say how many shares Anthropic will sell or what price they will carry. NPR called the offering an IPO that could break records. Those basics will come in an amended S-1 with a price range, which typically lands before the roadshow, the investor marketing period that precedes pricing. Until then, any valuation figure is a report, not a company disclosure.

What the S-1 says about revenue and customers

The headline figure is strong. Anthropic’s first-half 2026 revenue was $16.3 billion, up about 1,300% from $1.2 billion a year earlier, and the filing shows a first operating profit in the second quarter, according to The Next Web.

Most of that money comes from usage. Consumption-based revenue reached $13.1 billion, or 80% of the total, for the six months to June 30, CNBC reported. The company had more than 11,900 customers generating $100,000 or more in trailing-twelve-month revenue as of Sept. 30, 2026, per SQ Magazine, and it employed 5,300 people.

The third-quarter figure is missing. The filing leaves Q3 revenue blank. The Financial Times reported earlier that Anthropic told investors it reached $16.8 billion in the quarter, but that number is a report, not a filed result. Treat it that way until a later filing confirms it.

Another number readers will see is run-rate revenue, which was reported above $47 billion in May. A run-rate annualizes a recent period, so it cannot be compared directly with the six-month total. Keep the two figures apart when you judge growth.

Why the $80.7 billion loss is mostly an accounting story

Anthropic reported an $80.7 billion net loss for the six months ended June 30. A $72.4 billion fair-value increase on convertible notes, derivative liabilities and SAFEs accounts for most of that total, according to Tech Insider's breakdown of the filing. SAFEs, or simple agreements for future equity, are instruments startups use to raise money before a priced round.

The filing ties that jump to the company’s own valuation rising. Under fair-value accounting, when a company’s implied value climbs, the liabilities tied to shares it may issue later climb too, and that increase shows up as a loss. That is how a company can report an operating profit in one quarter and still post a loss this large.

A headline loss can mislead. Investors who want to judge the business should weigh revenue growth and the operating result over the net figure. The loss could also shift when the offering price is set, because the value of those instruments depends on the valuation.

Compute deals, control and the risks the filing flags

The filing lists long-term compute commitments of at least $114.5 billion with Google, at least $110 billion with Amazon and at least $44.9 billion with Microsoft. Those floors add up to at least $269.4 billion, which shows how much of Claude’s capacity depends on three cloud providers. Big Tech’s $3 trillion AI spending pledges reflect the same kind of bet, and the Anthropic figures read best alongside them.

That concentration cuts two ways. It secures capacity at a scale few AI labs can match. It also ties Anthropic’s growth to companies that sell their own AI models, Google among them. Business customers will want to know how these contracts affect pricing and availability, and the filing gives floors rather than full terms.

Ownership is the other issue. Seven co-founders will hold 50.1% of the vote through a single share class, according to reporting on the filing. Public buyers would enter a company where that small group holds a majority of votes. Anthropic is also a public benefit corporation, a structure that lets its board weigh the public interest alongside shareholder returns.

The filing’s risk language warns that “no single insurer” may be able to cover an AI catastrophe, according to coverage of the document. It also warns that Claude’s models could harm patients as the company expands into healthcare, MobiHealthNews reported. Healthcare sales bring liability and insurance questions that consumer chatbots mostly avoid, so those passages deserve a close read.

Anthropic IPO date: what the timing reports say

No date is confirmed, and the trackers do not agree. Tastytrade reports a roadshow as early as the week of Nov. 9 and a debut before Thanksgiving, which it puts at Nov. 26. Motley Fool points to marketing starting the same week. Leverage Shares reports a listing in the week of Nov. 16, and Decode the Future reports possible pricing on Nov. 19 or 20.

There is a timing problem. Leverage Shares says a roadshow cannot begin until at least 15 days after the public filing. Counting 15 days from Oct. 26 lands on Nov. 10, one day after the earliest roadshow start some reports give. Either the early date is wrong or the reports count from a different starting point, so check the rule against SEC guidance before planning around any single date.

The filing history has a gap too. Anthropic first filed confidentially in June. One tracker gives June 1, while another says only June, so the exact day remains unconfirmed.

Anthropic IPO price range and valuation

Reports put a potential valuation between $1.5 trillion and $2 trillion, a range NPR and Motley Fool have both described. The filing itself gives no target. Motley Fool also says Anthropic may aim for a valuation significantly below the roughly $2 trillion previously expected, so the top of that range is one scenario, not a forecast.

Against the $965 billion valuation from the company’s May 2026 Series H round, the reported range sits roughly 1.6 to 2.1 times higher. That is a steep step-up in a short period. Whether a weak market could delay the offering is also unresolved in public reporting.

The closest large precedent is SpaceX. Its June 2026 IPO priced at $135 a share and raised $86 billion at a valuation of about $1.77 trillion. By late July, its shares had fallen about 50% from their post-IPO peak, according to Wikipedia’s entry on the offering.

That history matters because investors will use it to judge the Anthropic deal. A large valuation at pricing does not guarantee the first months of trading. Revenue growth is the strongest case for a valuation this size, but a debut near $2 trillion would require buyers to bet that growth continues for years. For a sense of how the market rewards scale at the top end, see Apple’s $5.03 trillion market value.

What to do before ANT trades

If you plan to buy, the next few weeks offer more useful checks than any price target. Start here:

• Read the filing itself. The S-1 is on SEC EDGAR. Start with the risk factors and financial statements rather than summaries.
• Wait for the amended S-1. The price range, share count and offering size will come from that document, so forming a value view before then means guessing.
• Be wary of pre-IPO access. SQ Magazine notes that the filing period has drawn warnings about unofficial pre-IPO shares. Be skeptical of tokens or brokers that sell Anthropic exposure before it lists.
• Ask about allocations. Retail access to new listings is often limited, so ask your broker how it allocates IPO shares and whether you can order at the offering price.
• Confirm the exchange. CNBC notes that Anteros Metals, a Canadian land-exploration company, uses the same ANT symbol on a Canadian exchange, so verify the Nasdaq listing before you trade.

The next concrete signals will be an amended S-1 with a price range and share count, a roadshow start date and the first day of trading. Each one will show whether the reported valuation holds. Watch for the third-quarter revenue figure as well, since the blank line in the current filing is the most visible gap in the numbers.