In the ever-evolving world of artificial intelligence, Anthropic has made a significant move by signing a $35 billion deal for cloud computing power with Lambda, a US-based startup backed by tech giant Nvidia. This agreement marks another milestone in Anthropic's quest for advanced computing resources, essential for developing and deploying cutting-edge AI models.
The six-year deal, confirmed by a source familiar with the agreement, will provide Anthropic with substantial cloud computing capabilities from Lambda, a company known for its AI cloud services. Located in San Jose, California, Lambda is set to leverage a data center built by Hut 8, a firm that specializes in bitcoin mining and data center construction. Specifically, Anthropic will tap into a facility in Texas, enhancing its computational prowess.
Nvidia, which has invested in both Anthropic and Lambda, will lease the data center from Hut 8. Lambda will then pay an undisclosed amount to Nvidia for access to this facility, as reported by the Wall Street Journal. This intricate arrangement underscores the complex web of partnerships and investments driving the AI industry forward.
This deal is part of a broader strategy by Anthropic to secure extensive computing resources. Earlier this year, the company signed a similar six-year agreement worth $45 billion with Nscale, a London-based cloud provider. These deals cumulatively amount to at least $135 billion in computing agreements for Anthropic, highlighting the intense competition and demand for advanced computing power in the AI sector.
The competition between Anthropic and its rival, OpenAI, is fierce, with both companies vying for customers and aiming for initial public offerings. The surge in demand for chips and computing power has not only driven up costs but also strained global supply chains. Companies like Apple have already increased prices for certain devices due to rising memory and chip costs, a trend that is expected to continue as the demand for advanced technology grows.







































