Anthropic inks deal with Google, Broadcom to secure AI chips; tops $30 billion run-rate

Anthropic announced on Tuesday that it has signed an agreement with Google and Broadcom to secure multiple gigawatts of next-generation tensor processing unit (TPU) capacity that will come online in 2027.

The TPUs will be used to train and deploy frontier Claude models, to meet the growing demand for the company's technology offerings from customers worldwide.

Alongside the deal announcement, the IPO-bound AI company also revealed that its run-rate revenue has surpassed $30 billion, up from $9 billion at the end of 2025, as demand for Claude continues to accelerate worldwide.

"We are building the capacity necessary to serve the exponential growth we have seen in our customer base while also enabling Claude to define the frontier of AI development,” the CFO of Anthropic, Krishna Rao, said in a blog post.

Broadcom's agreement with the two tech giants

In a separate exchange filing, Broadcom said that it has entered a long-term agreement with Google to supply future generations of the internet giant's TPUs, which are designed to power artificial intelligence in data centres.

Apart from that agreement, Broadcom and Google have also expanded their collaboration to give Anthropic access to about 3.5 gigawatts of TPU-based compute capacity, according to a Securities and Exchange Commission (SEC) filing and Anthropic's blog post.

Claude's rising customer demand

Anthropic said that during its Series G fundraising in February, more than 500 business customers were each spending over $1 million annually on its services, the company said in a blog post, adding that the number has now surpassed 1,000, doubling in less than two months.

The startup also said that the vast majority of the new compute will be sited in the United States, making this partnership a major expansion of the company's November 2025 commitment to invest $50 billion in strengthening American computing infrastructure.

“The partnership deepens our existing work with Google Cloud—building on the increased TPU capacity we announced last October—as well as our relationship with Broadcom,” it added.

The development comes at a time when Anthropic is facing an ongoing dispute with the US government since it insisted that its technology should not be used for mass surveillance or fully autonomous weapons systems. In March, the Trump administration labelled Anthropic as a “supply-chain risk,” a designation which the AI company later challenged in court.

According to multiple media reports, the US defense forces employed Anthropic's AI to conduct airstrikes on Iran, which also killed Supreme Leader Ayatollah Ali Khamenei.

Anthropic IPO details

Anthropic PBC may debut on stock exchanges as early as October, as the AI firm competes with rival OpenAI Inc to launch an initial public offering, according to a Bloomberg report.

The developer of the renowned Claude chatbot has held preliminary talks with Wall Street institutions regarding lead roles in a potential listing, Bloomberg reported earlier, citing people familiar with the situation.

Goldman Sachs Group Inc, JPMorgan Chase & Co, and Morgan Stanley are reportedly being considered for primary roles in the Anthropic and OpenAI offerings.

Anthropic’s IPO could target a valuation exceeding $60 billion, The Information reported previously. Meanwhile, during a $30 billion funding round co-led by MGX that concluded in February, Anthropic was valued at $380 billion.