Key Highlights
- Revenue run rate climbed to over $30 billion from $9 billion since late 2025
- Enterprise clients spending $1M+ annually exceed 1,000, representing a twofold increase since February
- Strategic agreement with Google and Broadcom secured for next-gen TPU chip access beginning in 2027
- Computing capacity totaling approximately 5 gigawatts secured through the collaboration
- Broadcom stock jumped over 3% in after-hours trading on announcement
The AI company Anthropic has witnessed its revenue run rate surge past the $30 billion threshold, representing a remarkable leap from the $9 billion figure recorded at 2025’s conclusion. The Claude AI developer attributes this explosive growth to dramatically increased market demand throughout 2026.
The startup now counts over 1,000 enterprise clients each contributing more than $1 million in annual spending. This milestone represents more than a doubling of high-value customers since February.
To sustain its rapid expansion trajectory, Anthropic has finalized a significant computing infrastructure agreement involving Google and Broadcom. The comprehensive arrangement provides Anthropic with access to approximately 5 gigawatts of total computing capacity over multiple years.
Approximately 3.5 gigawatts of this infrastructure will be delivered via Google’s proprietary tensor processing units (TPUs), manufactured by Broadcom. This component of the partnership launches in 2027.
Broadcom and Google have additionally executed a long-term supply contract for these chips extending through 2031. Under the agreement’s terms, Anthropic will also engineer and provide customized TPUs specifically for Google.
Broadcom CEO Hock Tan previewed this partnership during last month’s quarterly earnings presentation. He projected that Broadcom’s AI semiconductor revenues would exceed $100 billion in the coming year.
The overwhelming majority of this new computing infrastructure will be deployed on U.S. soil. Anthropic emphasizes this aligns with its November 2025 pledge to channel $50 billion toward American computing infrastructure investments.
Growth Persists Amid Government Dispute
Anthropic currently faces an ongoing legal battle with federal authorities. The Department of Defense classified Anthropic as a supply-chain risk after disagreements emerged concerning AI safety protocols.
The company has cautioned that this designation threatens billions in potential lost contracts. According to testimony from a company legal representative, more than 100 enterprise accounts reached out expressing concerns about maintaining their Anthropic relationships.
Nevertheless, the organization reports uninterrupted growth momentum. Chief commercial officer Paul Smith noted that certain clients actually admire Anthropic for “standing by its principles” when engaging with government entities.
Implications of the Broadcom Partnership for Semiconductor Industry
Google initially developed its TPU technology to enhance search engine performance. The chips have evolved into essential infrastructure for developing and deploying artificial intelligence applications.
Broadcom transforms Google’s conceptual chip architectures into production-ready components that proceed to fabrication facilities. This collaboration establishes Broadcom as a competitive alternative to Nvidia, which maintains market dominance in AI computing hardware.
Broadcom equity appreciated up to 3.6% during after-hours sessions following the disclosure. Google parent company Alphabet gained approximately 1.6% in premarket activity, while Nvidia experienced a modest 0.4% decline.
Anthropic competitor OpenAI executed comparable computing agreements last year involving Broadcom, Nvidia, and additional suppliers to guarantee AI infrastructure availability.
Broadcom noted in regulatory documents that Anthropic’s utilization of the enhanced computing resources hinges on the company maintaining its commercial momentum.


