The AI race just got a lot more interesting. Anthropic, the company behind the popular Claude AI, has filed for its initial public offering (IPO), giving us the first official look at the financials powering one of OpenAI’s biggest rivals.
The Headline Numbers: Huge Growth, Deeper Losses
Anthropic's
filing reveals a company experiencing explosive growth. For the first nine months of 2026, revenue soared to $33.2 billion, a staggering 13-fold increase from the same period in 2025. The third quarter alone brought in $17 billion. However, this growth comes at a tremendous cost. The company reported a net loss of $104 billion for those nine months. This dynamic of rapid revenue growth paired with significant losses is common for ambitious tech companies in a high-stakes, high-investment phase, but the scale is notable even for the capital-intensive AI sector.
How Anthropic Makes Its Money
The company’s revenue model is primarily driven by usage of its Claude family of AI models. The filing shows that about 79% of its revenue comes from consumption-based charges, where customers pay for the amount of data processed by the AI. The remaining 21% comes from subscriptions for its various products like Claude Pro and enterprise-level plans. A significant portion of this business, about 44% of revenue, flows through the marketplaces of its cloud partners like Google and Amazon—who are also, in a complex twist, some of its biggest competitors.
The Staggering Cost of Cutting-Edge AI
The massive losses are explained by one core factor: the colossal cost of computing power. To train and run advanced AI models, companies like Anthropic need vast fleets of specialized servers and GPUs. The filing disclosed that Anthropic has committed to spending a jaw-dropping $895.4 billion on cloud computing and infrastructure over the coming years. Of that, nearly $70 billion is due within the next 12 months. These expenses for training and 'inference' (the cost of running a query) are the primary reason why profitability remains elusive, despite impressive revenue growth.
The 'Frenemy' Factor: Partners and Rivals
Anthropic's IPO filing sheds light on its complicated relationships with tech giants. It has massive spending commitments with Amazon and Google, who are also major investors in the company through convertible notes. At the same time, Google's Gemini and models from Microsoft-backed OpenAI are Anthropic's primary competitors. This means Anthropic is effectively buying the essential tools for its business from its direct rivals, a situation the filing identifies as a fundamental risk. The IPO is a clear move to raise independent capital and reduce its reliance on these dual-natured partners.
What the Future Holds
The company's strategy, as outlined in the filing, is to use the IPO proceeds to fuel further research and development, expand its infrastructure, and grow its enterprise customer base. The filing proudly notes that over 11,900 customers generated at least $100,000 each in the past year. Anthropic is betting that its focus on building safer and more steerable AI will be a key differentiator in the enterprise market. However, the filing also lists 80 pages of risk factors, highlighting the intense competition, regulatory uncertainty, and the relentless pace of innovation as major challenges.
