Anthropic's Growth Comes With a Nearly $900 Billion Infrastructure Commitment

Anthropic's growing enterprise business is generating billions in revenue, while its latest IPO filing reveals the enormous infrastructure commitments required to sustain its expansion.

[For more news, click here]

Revenue reached $16.96 billion in the third quarter of 2026, compared with $1.38 billion in the same period last year, as demand for Claude continues to grow across businesses. Alongside its financial results, the company disclosed $895.4 billion in future purchase and lease commitments, largely related to computing infrastructure. The scale of those obligations gives a sense of how much capital is now required to compete in the advanced AI market.

The figures were disclosed in an amended registration statement filed with the US Securities and Exchange Commission on November 2, as Anthropic prepares for its proposed public listing. While its revenue reflects growing commercial adoption, the filing also shows the financial pressures involved in securing enough computing capacity to develop and operate its models.

Enterprise adoption drives Claude's growth

Businesses are becoming an increasingly important source of revenue for Anthropic. The company reported more than 11,900 customers generating at least $100,000 in revenue over the preceding 12 months as of September 30, compared with more than 950 a year earlier. Software development, research and business operations are among the areas where companies are incorporating Claude into their products and internal workflows.

However, attracting customers does not automatically translate into lower operating costs. Anthropic reported an operating loss in the third quarter, partly because it had to secure computing capacity through short-term agreements that were more expensive than comparable capacity under longer-term arrangements. As demand grows, the company must continue investing in the infrastructure needed to serve customers while managing the cost of that capacity.

The cost of securing AI infrastructure

Anthropic's $895.4 billion in future purchase and lease commitments covers several years, including computing capacity and facilities that have yet to become operational. These are contractual obligations rather than money already spent. A substantial part of the commitments relates to long-term arrangements for data center capacity and the equipment needed to run advanced AI systems.

The company is also developing Anthropic Direct, an approach that involves leasing chips and placing them in dedicated facilities operated on its behalf. Rather than relying entirely on conventional cloud agreements, this allows Anthropic to arrange data center capacity separately from the computing equipment installed inside those facilities.

Major financial institutions are also taking a role in funding the infrastructure. On October 26, Broadcom, Bank of America, Blackstone, Citi and Morgan Stanley announced a $63 billion financing arrangement for AI accelerators and networking equipment that will be leased to Anthropic, with deliveries expected to begin in 2027. The equipment will be acquired through a special-purpose vehicle, with the banks and investment firms providing financing and Broadcom supplying the hardware.

For semiconductor manufacturers and financial institutions, AI infrastructure is opening up business beyond the development of models and software. Broadcom supplies the specialized chips and networking equipment, while the participating financial institutions are financing the hardware acquisition. It is an indication of how much of the AI industry's expansion now depends on companies and investors outside the model-development business.

What investors will be watching ahead of the IPO

Anthropic's proposed listing will bring greater scrutiny to its business model. Its enterprise customer base and revenue growth provide a strong commercial foundation, but investors will also be assessing whether the company can manage its infrastructure commitments without allowing operating costs to outpace the returns from its AI services.

The company has applied to list its Class A common stock on the Nasdaq Global Select Market under the ticker symbol ANT. Its registration statement also outlines its public benefit corporation structure, under which it considers its stated purpose of developing advanced AI responsibly alongside shareholder interests. Its Long-Term Benefit Trust will retain an important role in governance, including certain board appointment responsibilities.

Anthropic has established a substantial business around demand for its AI models. Its next challenge is managing the infrastructure required to support that business while maintaining a path toward sustainable profitability. The company's public listing will give investors a clearer view of how it intends to balance those priorities.

Related Articles

Broadcom’s $63 Billion Anthropic Deal Brings Wall Street Deeper Into AI Infrastructure

Google Puts AI Chips in Orbit to Test a New Kind of Data Center

VAST Data and AMD Expand AI Infrastructure Alliance to Tackle the Industry's Inference Bottleneck