Anthropic Has Billion in Cloud Bills and No Credibility Left to Warn About AI

Anthropic has committed to spend $200 billion with Google Cloud over five years, The Information reported Tuesday, citing a person with knowledge of the matter. The commitment makes Anthropic the largest single customer in Google Cloud's pipeline, representing more than 40 percent of the revenue backlog Google disclosed to investors last week, according to Reuters.

It also means Anthropic, the company that has positioned itself as the industry's independent conscience, cannot warn the world about the dangers of AI concentration without warning about itself.

The deal is the latest layer in an infrastructure stack Anthropic has been assembling since 2025. In April, the company signed a deal with Google and Broadcom for 3.5 gigawatts of next-generation TPU capacity, coming online in 2027, according to Anthropic's own announcement. That same month, Amazon disclosed it was investing up to $25 billion in Anthropic, and Anthropic pledged more than $100 billion in spending on AWS technologies over ten years, Anthropic reported separately. Alphabet is separately investing up to $40 billion in Anthropic, $10 billion now at a $350 billion valuation, with another $30 billion contingent on performance targets, TechCrunch confirmed.

Contracts involving Anthropic and OpenAI now account for more than half of the $2 trillion in backlogs across Amazon Web Services, Microsoft Azure, and Google Cloud Platform, The Information's report said. Alphabet shares rose about 2 percent in extended trading following the news.

Anthropic's run-rate revenue has surpassed $30 billion, up from approximately $9 billion at the end of 2025, the company disclosed in April. More than 1,000 business customers now spend more than $1 million annually with Anthropic, doubling that cohort in less than two months.

Anthropic framed the infrastructure push as necessity, not strategy. "We are committing more than $100 billion over the next ten years to AWS technologies," Amodei wrote in the Amazon announcement, noting the company needed compute to keep up with demand. With run-rate revenue above $30 billion, the spending is also rational infrastructure planning by a company that can afford it. But it is also a spending decision that has consequences for what Anthropic can credibly say publicly.

The $200 billion figure is not a surprise given the scale of what Anthropic has been building. But it crystallizes a tension the company has managed largely by avoiding it: an AI lab that publishes safety research and warns about frontier model risks, now financially locked into the infrastructure layer of two companies it competes with in AI applications. Anthropic declined to comment. Google redirected queries to Anthropic.

The deal begins next year, according to The Information. It will fund the TPU capacity Anthropic needs to train and serve Claude at a scale that currently exceeds what its existing infrastructure can reliably handle.

Anthropic has spent years publishing safety research and essays on the risks of AI concentration. In a 2024 essay on AI development, Amodei wrote that "the most responsible path" for frontier labs was to maintain independence from the companies whose behavior they were studying. The $200 billion commitment to Google Cloud does not break that promise in any legal sense. It makes the promise economically incoherent. When a company spends more than most governments on its competitor's infrastructure, its published warnings about that competitor carry a footnote the reader can do the math on.

The practical effect is not hypothetical. Anthropic has published research on the risks of AI concentration, including the danger of a small number of companies controlling the compute layer that underlies the entire industry. That research is now authored by a company that is also one of the largest financial commitments in the history of cloud computing to one of the companies it is warning about. Enterprise buyers and regulators who cited Anthropic's safety posture as a reason to trust the company are left to reassess.

What to watch next: whether Anthropic's published safety posture adjusts to match its financial anatomy, or whether the independence framing simply ages quietly into the company's folklore.