Anthropic’s IPO: A close look at AI frontrunner’s rapid growth, spiraling costs

By Deborah Mary Sophia

Oct 26 (Reuters) - Anthropic unveiled its US initial public offering paperwork on Monday, setting up for a blockbuster market debut that would test investors’ willingness to fund the AI juggernaut’s bills amid growing worries over the massive debt-fueled spending.

The company had submitted a confidential filing in June. In the recent weeks, exclusive reporting from Reuters highlighted its sweeping AI ambitions, mounting costs and concerns about the "existential risks to humanity" that advanced AI poses.

Here are a few details on Anthropic’s financials, business structure, and where it stands in the global AI race:

BACKGROUND AND FINANCES

Founded in 2021 by former OpenAI researchers including siblings Dario and Daniela Amodei, Anthropic is the startup behind the Claude family of AI models. Across four tiers - Fable, Opus, Sonnet and Haiku - Claude powers products such as Anthropic’s chatbot, coding agent and the enterprise-focused Claude Cowork AI agent.

Anthropic’s early focus on training its models for coding tasks propelled the company to the forefront of the AI race and helped it draw a steady stream of enterprise customers.

A majority of its revenue comes from businesses paying for its models per use, measured in "tokens" of information processed, in what Anthropic calls "consumption-based" revenue. It also generates revenue from its paid customer subscription tiers such as "Pro" and "Max".

That has helped Anthropic grow its revenue to $4.59 billion in 2025 from $386 million a year ago and $41 million in 2023. For the six months ended June 2026, Anthropic’s revenue was $16.29 billion.

But the rapid growth is accompanied by bigger losses. Its operating loss for the first half of 2026 was $693 million. In 2025, it was $8.06 billion, up from $2.98 billion a year ago.

COSTS, COMPUTE DEMAND AND CIRCULAR AI INVESTMENTS

Computing power is Anthropic’s biggest expense.

The company’s costs include computing fees paid to cloud services providers such as Amazon Web Services and Google Cloud, costs tied to data center facilities and renting data center capacity, equipment leases and energy bills.

Over the past three years, Anthropic’s compute and infrastructure expenses have dwarfed its revenue.

The company has committed more than $500 billion in spending with ​six partners over a decade - one of the largest AI buildout commitments ‌on record. Anthropic said many portions of its agreements are non-cancelable or require payments regardless of usage.

It has committed to spend at least $110 billion with Amazon, about $114.5 billion with Google and $44.9 billion with Microsoft for cloud computing over the next seven to 10 years.

But some of those companies are also its largest backers. Amazon and Google are among Anthropic’s top shareholders, marking yet another example of the circular financing structures that have dominated capital flows in the AI industry.

Amazon has invested at least $13 billion in Anthropic and has agreed to invest up to $20 billion more. Google has committed up to $40 billion to Anthropic.

GOVERNANCE AND STRUCTURE

Anthropic is a public benefit corporation, or PBC, meaning it is legally obligated to balance profitability goals for investors with developing AI for the benefit of humanity. An independent body elects a majority of its board.

The company, helmed by Dario Amodei as CEO and his sister Daniela, as president and chair of the board, will continue to operate as a PBC after its listing.

It expects to keep its founders and an independent trust in control, in a structure that could limit investor oversight and public shareholders’ influence. The company has said it needs to concentrate control and capital in fewer hands to reduce AI’s risks.

Under this arrangement, a majority vote of its seven co-founders will direct a single share of Class F stock that amounts to 50.1% of the total voting power over key corporate matters, including the election of some board directors and other matters submitted to investors.

COMPETITION

Anthropic’s most direct competitor is OpenAI, the creator of ChatGPT, whose 2022 launch kick-started the global race to advance and adopt AI technologies.

Following Anthropic’s lead in the enterprise market, OpenAI has also sharpened its focus on products for developers and business customers.

Anthropic is also locked in a race with Elon Musk’s SpaceX and Alphabet’s Google. All of them are spending billions of dollars into model development, data centers and specialized AI chips as they vie for dominance in generative AI.

Ramp data, which captures a slice of AI spending by business customers, showed Anthropic accounted for 60.1% of AI token spending in the week ended October 11. The startup has gained ground against rivals increase spending on AI services.