Anthropic’s Loss Narrows to $23.3 Billion as Revenue Jumps Ahead of Record-Breaking IPO

Rows of server racks in a data centre Photo: NERSC / Wikimedia Commons (CC0)

Anthropic, the artificial intelligence company behind the Claude chatbot, has reported a sharply narrower quarterly loss alongside a jump in revenue, in an amended filing that sets the scene for a stock market listing that could be the largest on record.

The company’s net loss fell to $23.29 billion in the three months to 30 September, down from $66.3 billion in the second quarter, according to the filing with the US Securities and Exchange Commission published on Monday. Revenue rose to $16.96 billion over the same period, up from $11.56 billion in the previous quarter.

The quarterly improvement does not mean the business is close to profitability on a headline basis. The third-quarter loss was still wider than the $12.1 billion recorded a year earlier, and operating expenses jumped 292 per cent year on year to $17.5 billion. Anthropic swung from an operating profit of $974 million in the second quarter to an operating loss of $567 million in the third, the filing showed.

For the nine months to the end of September, revenue reached $33.24 billion, while the net loss widened to $104 billion, compared with $25.1 billion in the same period a year earlier. The scale of those losses reflects, in large part, the enormous cost of building and running the computing systems behind advanced AI models, including heavy stock-based pay.

That cost is laid bare elsewhere in the filing. Anthropic disclosed about $895.4 billion in purchase and lease commitments, substantially all of which relate to computing power and infrastructure. The figure underlines how much capital the leading AI laboratories are locking in for chips, data centres and energy before they have turned a sustained profit.

Anthropic filed publicly for an initial public offering on 26 October, after first filing confidentially in June. It plans to list on the Nasdaq Global Select Market under the ticker symbol ANT, with Goldman Sachs and Morgan Stanley leading the offering alongside 20 other banks, according to the filing. The number of shares and the price range have not yet been set.

Reports, rather than the filing itself, have put the potential scale of the deal in context. People familiar with the matter have told Bloomberg News that Anthropic is seeking to match or exceed the $86.2 billion raised by SpaceX in its market debut, and that investors have indicated a fair valuation of around $1.4 trillion to $1.5 trillion, with a figure closer to $1.75 trillion potentially achievable. Those terms are reported, not confirmed in the filing, and the listing is expected to take place before the US Thanksgiving holiday.

The filing lands at a moment of intense scrutiny of AI spending. Investors have spent recent weeks questioning whether the vast sums being poured into AI infrastructure will generate returns quickly enough to justify current valuations, a debate that helped drive a sharp sell-off in technology shares in October.

Note: the featured image is a file photo of server racks in a data centre, not a photograph of an Anthropic facility.

Sources: Reuters, “Anthropic’s quarterly loss shrinks, revenue jumps ahead of blockbuster IPO”, 2 November 2026, https://www.reuters.com/business/anthropics-quarterly-loss-shrinks-revenue-jumps-ahead-blockbuster-ipo-2026-11-02/ ; Bloomberg News, “Anthropic Revenue Hits Nearly $17 Billion Before Mega-IPO”, 2 November 2026, https://www.advisorperspectives.com/articles/2026/11/02/anthropic-revenue-17-billion-before-mega-ipo

Leave a Reply

Your email address will not be published. Required fields are marked *