Anthropic has told shareholders it expects positive adjusted operating income for a second straight quarter, the Financial Times reported, citing several people with knowledge of the matter. CNBC reported the same the next day, citing two people familiar with the matter. The news comes as the Claude maker prepares for an initial public offering on Nasdaq, widely expected as soon as October, The Next Web reported.
The numbers, and what they leave out
Anthropic’s quarterly revenue passed $11.5 billion in the second quarter, with positive adjusted operating income. Its annualized revenue reached $65 billion in July, about seven times the level of a year earlier, according to CNBC. The FT reported that gross margins are above 80%.
Both measures leave out large costs. The margin figure comes before the revenue Anthropic shares with distribution partners, including Amazon, and before the cost of training its models. The adjusted profit measure strips out costs that Bloomberg described as exceptional or one-time items. Anthropic hasn’t published full accounts.
IPO on track
Anthropic filed its IPO prospectus confidentially in June. It must make its financials public at least 15 days before its investor roadshow starts, Business Insider reported. Bloomberg reported that the company aims to match or beat SpaceX’s record $86.3 billion offering and is finalizing a $15 billion revolving credit facility.
Matt Murphy, a partner at Anthropic investor Menlo Ventures, called the company’s growth “off the charts.” “Don’t see why growth would slow or any other reason to wait,” he told CNBC.
The safety backdrop
The profit news landed in a week of AI safety warnings. CEO Dario Amodei had just called on the industry to slow down. OpenAI’s Sam Altman agreed and said his company would not go public this year. Multiple sources told Axios’ Dan Primack that Anthropic is still likely to list in 2026. Primack also noted a “cynical argument” that slowing development could cut Anthropic’s compute spending and improve its financials.
Not everyone is convinced. D.A. Davidson analyst Gil Luria told CNBC he is “highly suspicious” of what Anthropic and OpenAI are doing. “It feels more and more like a ladder pull,” he said. Gartner’s Arun Chandrasekaran said stricter safety standards could favor the two labs if smaller rivals can’t afford the costs. Deepwater Asset Management’s Gene Munster expects little to change. “AI’s long-term opportunity is too big for them to slow down,” he said. The public filing will be the first real test, because it will show how the adjusted figures compare with GAAP results once training costs and partner payments are included.
WebProNews is an iEntry Publication