TLDR
- Anthropic’s annualized revenue has climbed beyond $30 billion from $9 billion recorded at the close of 2025
- More than 1,000 enterprise clients now invest over $1 million each year, representing a 100% increase since February
- A strategic agreement with Google and Broadcom secures next-gen TPU chip infrastructure beginning in 2027
- The arrangement provides Anthropic with approximately 5 gigawatts of total computing capacity
- Broadcom stock gained over 3% after the partnership disclosure
The AI company Anthropic has seen its revenue run rate surge from $9 billion at 2025’s conclusion to exceeding $30 billion, marking a dramatic expansion confirmed this week. The startup attributes this growth to rapidly increasing demand for its Claude AI platform throughout 2026.
The company now serves over 1,000 corporate clients each contributing more than $1 million annually. This figure has experienced a remarkable doubling since February of this year alone.
In response to escalating demand, Anthropic has finalized a substantial computing partnership with Google and Broadcom. The arrangement will deliver approximately 5 gigawatts of total computing infrastructure in the years ahead.
Approximately 3.5 gigawatts of this capacity will be supplied via Google’s tensor processing units (TPUs), manufactured by Broadcom. This segment of the partnership is scheduled to launch in 2027.
Broadcom and Google have also established a long-term supply contract for these chips extending through 2031. Additionally, Anthropic will contribute to developing and providing customized TPUs for Google under this arrangement.
Broadcom CEO Hock Tan referenced this partnership during a recent earnings conference call. He projected that Broadcom’s AI chip revenue would exceed $100 billion in the coming year.
The bulk of this new computing capacity will be deployed within the United States. Anthropic notes this expansion supports its November 2025 pledge to direct $50 billion toward American computing infrastructure development.
Legal Challenges Haven’t Derailed Momentum
Anthropic is currently engaged in a legal confrontation with federal authorities. The Pentagon designated Anthropic as a supply-chain security concern after disagreements emerged regarding AI safety protocols.
Anthropic has cautioned that this designation could result in billions of dollars in forfeited revenue. According to a company legal representative, over 100 enterprise clients reached out expressing concerns about maintaining their partnerships with Anthropic.
Nevertheless, the organization reports sustained expansion. Chief commercial officer Paul Smith indicated that certain customers appreciate Anthropic’s willingness to “demonstrates its principles” when engaging with government entities.
Implications of the Broadcom Partnership for Semiconductor Industry
Google’s TPUs were initially developed to enhance search engine performance. They have evolved into essential infrastructure for developing and deploying AI applications.
Broadcom transforms Google’s chip specifications into complete products ready for manufacturing. This arrangement establishes Broadcom as a competitive alternative to Nvidia, which currently leads the AI computing market.
Broadcom shares increased by as much as 3.6% during after-hours trading following the announcement. Alphabet, Google’s parent company, rose approximately 1.6% in premarket activity, while Nvidia dipped 0.4%.
Anthropic’s competitor OpenAI established comparable computing partnerships last year with Broadcom, Nvidia, and additional suppliers to guarantee AI infrastructure access.
Broadcom noted in its regulatory filing that Anthropic’s utilization of the enhanced computing resources is contingent upon the company’s ongoing commercial performance.
