Nvidia plans to invest up to $100B in OpenAI

In a groundbreaking announcement on Monday, Nvidia revealed plans to invest as much as $100 billion in OpenAI, aimed at significant expansion of data centers dedicated to training and executing AI models. This partnership, formalized through a letter of intent, includes an ambitious plan to deploy 10 gigawatts of Nvidia systems, which is sufficient to power millions of homes, tailored specifically for OpenAI’s upcoming AI infrastructure.

This substantial investment could offer OpenAI a greater degree of independence from Microsoft, its primary investor and cloud service provider. Earlier in January, Microsoft had declared a shift in its partnership with OpenAI, permitting the AI powerhouse to explore additional infrastructure collaborations with other tech entities. Since then, OpenAI has forged alliances for various AI data center initiatives, notably the transformative Stargate project, in collaboration with others like Oracle and SoftBank.

As part of this new strategic relationship, OpenAI will consider Nvidia as its “preferred strategic compute and networking partner” for the growth of its AI manufacturing capabilities. Details regarding the structure of Nvidia’s investment—whether it be in terms of hardware, cloud credits, cash, or a combination—remain to be disclosed.