Anthropic’s annualized revenue surpasses $65 billion; the craziest AI story may no longer be models, but the speed of monetization

As of the end of July, Anthropic’s annualized revenue run rate had exceeded $65 billion, compared with just $47 billion in May and approximately $9 billion at the end of 2025. In other words, in seven months, it increased its revenue run rate to more than seven times its original level.
In a word: Claude is no longer just a “chatbot that many people find useful”; it is becoming a productivity tool that enterprises are genuinely willing to keep paying for. Anthropic’s actual second-quarter revenue exceeded $11.5 billion, up approximately 14-fold year over year.
But this is the easiest point to misread: $65 billion does not mean it has already collected $65 billion this year; it is the full-year run rate extrapolated from its most recent revenue pace. The truly astonishing part is not the absolute figure, but that growth is still accelerating on such a large base.
This data is important for the entire AI market because it is beginning to answer a question debated for two years: Can real demand actually absorb the hundreds of billions of dollars being poured into GPUs, data centers, and electricity? Anthropic has at least demonstrated that enterprises are indeed paying at scale.
But the opposing view is also straightforward: surging revenue does not mean profits are surging at the same rate. Model inference, training, computing power, and talent costs are all extremely high, and it remains unproven whether AI customers will stay locked into one model company over the long term or simply switch to whichever provider is cheaper.
Scenario A: If Anthropic can maintain this growth rate through the end of the year, the market will continue to revalue the entire AI industry chain, making capital expenditures on GPUs, memory, cloud services, and data centers easier to justify as “backed by revenue.”
Scenario B: If revenue growth begins to turn down significantly, or a price war erodes profits, today’s extremely high AI valuations could instead become the biggest source of pressure.
What matters more now is not whether Claude or GPT ranks first, but who can truly turn “model capabilities” into sustained cash flow. The $65 billion figure shows that the AI war has moved from competing on parameters to competing on commercialization. #Anthropic年化营收突破650亿美元