Anthropic’s IPO Filing and Financial Results

Anthropic has officially filed for an initial public offering, bringing its financial performance into public view as the company prepares to enter the stock market. The filing reveals a business growing rapidly while also recording mounting losses.

Revenue reached $5.6 billion in 2025, up from $1 billion in 2024. The company’s annualized revenue run rate had surpassed $30 billion by the end of March 2026, according to the filing. That rapid growth reflects the scale of Anthropic’s expansion, but it has not yet translated into profitability.

Anthropic reported a net loss of $5.2 billion for 2025, compared with $5.6 billion in 2024. Its expenses include substantial investment in computing capacity and infrastructure, as well as costs associated with developing its AI systems. The filing also notes that rising demand has made access to computing resources a key constraint on the company’s ability to serve customers and grow.

Together, the figures present a striking contrast: revenue is surging, while losses remain substantial. Investors will weigh that growth against the cost of sustaining it as they assess Anthropic’s IPO.

Enterprise Adoption

The filing shows that Anthropic’s products have gained significant traction with large businesses. By the end of 2025, more than 500 customers were spending over $1 million annually, while more than 40,000 customers were spending over $100,000 per year. These figures indicate that adoption extends beyond experimentation to substantial, recurring use across organizations.

Anthropic’s customer base also spans a broad range of industries, including financial services, healthcare, technology, and retail. The filing points to enterprise use of Claude for tasks such as software development, customer support, and research. It also describes growing demand for Claude Code, its coding product, and for Claude models accessed through cloud providers and other platforms.

This breadth matters because businesses can adopt Anthropic’s products through several routes, including direct subscriptions and third-party services. The filing’s customer-spending figures provide evidence of how far that adoption has progressed. At the same time, they do not mean every customer uses Claude in the same way: organizations can apply the models to different workflows and scale their use at different rates.

Governance and Founder Control

Anthropic’s governance structure is designed to give its founders unusually durable influence over the company, even as it grows and potentially prepares for a public offering. Rather than relying only on conventional shareholder voting, the arrangement gives founders a distinct role in selecting members of the board. That structure is intended to make it harder for outside investors or future public-market shareholders to redirect the company’s priorities simply by accumulating voting power.

The design reflects a tension at the heart of Anthropic’s corporate model: the company needs capital and commercial growth, while its founders want to retain meaningful control over decisions. The governance framework gives the founders a way to protect their influence without requiring them to own a majority of the company’s economic interests. It also establishes a formal mechanism for continuity in board composition, rather than leaving control entirely to ordinary shareholder elections.

This unusual arrangement is part of how Anthropic seeks to preserve its independence as it expands. It does not eliminate the influence of investors or other stakeholders, but it gives founders a stronger voice in the company’s oversight than a standard governance structure would. That influence could matter as Anthropic balances business demands with its stated priorities.

Protecting Anthropic’s AI Safety Mission

Anthropic’s governance structure is designed not only to determine who controls the company, but also to help protect its AI safety mission as it grows. The structure includes a Long-Term Benefit Trust, which has no economic interest in Anthropic but holds voting rights. Its trustees are independent of the company and are intended to represent the interests of humanity in its long-term decisions.

The trust’s role is expected to expand over time. It will eventually appoint a majority of the members of Anthropic’s board, giving it meaningful influence over the company’s direction. The board also includes a majority of independent directors, and the trust has a role in selecting them. These arrangements are intended to make the company’s governance more accountable to its stated purpose, rather than relying solely on the preferences of investors or executives.

This approach reflects the challenge Anthropic faces as it seeks to raise capital and pursue commercial growth while maintaining its focus on building AI systems safely. The trust does not run day-to-day operations, but its voting rights and future board appointments are intended to help preserve the mission over the long term.