🤖 Anthropic had an $89.7 billion net loss in H1 2026 🤪

21 Oct, 2026

Anthropic is generating breakneck revenue growth but faces a long list of risks that include a narrow customer base and uncertain access to much-needed computing power as it prepares for a mega-IPO, according to new disclosures.

The company has signaled that it could potentially spend $750 billion in the coming years on obtaining more computing capacity, including deals it is still negotiating.

The disclosures, many of which have not previously been reported, are part of a draft initial public offering prospectus viewed by The Wall Street Journal that revealed the firm’s financials for the first half of the year. Anthropic declined to comment.

The artificial intelligence giant said in the draft prospectus it recorded a net loss of more than $80 billion in the first half of this year, a total that largely reflects an accounting adjustment related to financing agreements.

That compares to a net loss of $13 billion during the same period a year earlier.

...

Anthropic’s revenue growth is closely tied to a few big customers. The firm said around one-fifth of its $16.3 billion in revenue in the first half of 2026 came from just two customers. The company expects a “significant portion” of its revenue will come from a “limited number of customers for the foreseeable future,” according to the draft prospectus.

Here are some of the biggest takeaways from the S-1.

Computing costs are growing fast—revenues are increasing even faster.

Anthropic generated around $16.3 billion in revenue in the six months ended June 30, up from $1.1 billion in the year-earlier period, according to the draft prospectus. During that period, Anthropic spent nearly $9 billion on computing capacity and infrastructure, up from $2.3 billion spent on infrastructure during the same six months in 2025.

Anthropic spent more on computing capacity in the first half of this year than it did in all of 2025, when it spent $7.3 billion and generated $4.6 billion in revenue.

Anthropic needs more computing power than the market can provide.

The company is on the hook to pay around $450 billion in cloud computing agreements over the next seven to ten years. The bulk of these deals are noncancelable.

Anthropic plans to spend $111 billion on Google cloud computing contracts, $110 billion with Amazon Web Services, and $31 billion on Microsoft, “regardless of actual usage.” The firm also agreed to spend up to $86 billion renting Nvidia chips from smaller AI cloud providers Nscale and Lambda through 2033 and 2034, respectively, in addition to other cloud deals.

Including its more than $160 billion obligation to lease Google chips in a deal with Broadcom, Anthropic’s future infrastructure costs top $600 billion. The firm is negotiating additional computing deals that could see its commitments rise by as much as $150 billion, the materials show.

While most of its capacity comes from large cloud providers—including Amazon Web Services, Google and Microsoft—Anthropic is hitting limits. The company said it has sought access to more capacity than “those partners were willing to provide” and they increasingly see it as a competitor, according to the prospectus.

...

Anthropic booked a massive loss driven by accounting charges.

The company recorded a net loss of $80.7 billion in the six months ending June 30, far more than the $42 billion it lost in all of 2025 and one of the largest half-year losses ever reported.

The vast majority of that loss, $72 billion, resulted from accounting charges tied to changes in the value of convertible notes and other instruments the company issued to raise funding.

Anthropic sees damages caused by AI as a giant financial risk.

Roughly a third of the draft prospectus was dedicated to risk factors, including the possibility of misaligned models causing real-world harms. Anthropic said AI models can show “self-preserving” qualities to avoid being shut down and can generate responses that resemble “blackmail” in an effort to convince humans to allow them to continue operating.

In some cases, Anthropic said it bears liability for certain harmful model behaviors, meaning it would be on the hook to pay for damages, such as stolen data or security incidents, caused by its models. But the company warned that “harm that could result from a catastrophic AI-related incident” might be so large that it would be unable to obtain insurance to cover the potential damages.