Anthropic's IPO Filing: $16.3 Billion in Six Months, an $80.7 Billion Loss, and a Trust That Picks the Board

On 26 October 2026, Anthropic published the document every company must file before selling shares to the public. It applied to list on Nasdaq as ANT. The S-1 shows a company whose revenue more than tripled in a year and whose losses ran far larger still, and a share structure built so that public investors cannot take control of it.

Source: Anthropic Form S-1, as reported by Yahoo Finance and CNBC.

Anthropic's Revenue and Losses in the S-1

The growth sits in the second quarter. Revenue was $4.7 billion from January to March 2026 and $11.6 billion from April to June. Over the same two quarters, the operating line swung from a $1.7 billion loss to $974 million of operating income.

The $80.7 billion net loss for the half is far larger than the operating figures; the 2025 net loss was $42 billion. Revenue per employee passed $6.6 million over the past year, against $1.4 million a year earlier, while headcount nearly tripled to about 5,300.

The customer base is concentrated at the top. As of 30 September, more than 11,900 customers had paid Anthropic $100,000 or more over twelve months, more than 1,900 had paid $1 million or more, and about 250 had paid $10 million or more. The filing warns that "a substantial portion of its revenue comes from a limited number of customers and distribution channels." Cash and short-term investments stood at about $131.7 billion on 30 June, from $141.1 billion raised since 2021.

Who Controls Anthropic After the IPO

Anthropic is a public benefit corporation. Its stated purpose, repeated in the filing, is "the responsible development and maintenance of advanced AI for the long-term benefit of humanity." The share structure protects that purpose from shareholders:

Buyers of ANT will own part of the revenue and none of the steering wheel.

The Risks Anthropic Lists

Most prospectuses warn about competition and regulation, and this one does: it flags "increasingly restrictive AI regulation." It also says the insurance industry as a whole may not be able to cover its liability from a catastrophic AI incident, and that AI could pose "catastrophic or existential risk to humanity." Few companies have listed shares while describing their own product that way.

Compute is the other large commitment. The filing describes arrangements worth up to $120.5 billion running into 2029, and a plan, called Anthropic Direct, to lease data centres and buy AI chips directly rather than rely only on Amazon and Google cloud capacity.

Anthropic IPO Date and Valuation

The share count, price range and third-quarter results were left blank, to be added in an amended filing before marketing begins. Based on the SEC timetable, Yahoo Finance put the earliest possible pricing date at 13 November and the earliest trading day at 16 November. Anthropic has not confirmed either.

The last private valuation was $965 billion, in May 2026. Press reports on the listing have cited $1.5 trillion to $2 trillion; those are not figures from the filing. Insider lock-ups expire in stages on days 27, 48, 62 and 76 after listing, so more shares can reach the market in the weeks after the debut.

Also readWhat does the Claude for Startups programme give founders?

For companies building on Claude, the S-1 adds one practical fact: Anthropic's model business now reports quarterly results, risks and customer concentration in public. Naraway's AI integration team builds on Claude and other models for products that need to run in production.