Just weeks after securing a monumental $100 billion agreement with Nvidia for computer chips, AI powerhouse OpenAI has now forged a similar partnership with Advanced Micro Devices (AMD).
This new alliance positions AMD as a key supplier for OpenAI, entering a market currently dominated by Nvidia and fueling the intense competition among chipmakers striving to power cutting-edge artificial intelligence technologies like ChatGPT.
OpenAI announced on Monday that it plans to integrate AMD chips into its operations starting in the latter half of next year. These chips will be deployed in new computer data centers, distinct from the facilities OpenAI is already developing in locations such as Texas, New Mexico, and Ohio, as well as a yet-to-be-named Midwest site.
The scale of this new deployment is immense: OpenAI anticipates utilizing enough AMD chips over the coming years to require 6 gigawatts of power. To put that into perspective, this is equivalent to the electricity needs of all households in Massachusetts. This follows a previous agreement with Nvidia, which involved deploying chips requiring 10 gigawatts of power.
While AMD is not making a direct cash investment in OpenAI, the terms of the agreement offer OpenAI the option to purchase up to 160 million shares in AMD at a mere penny per share. This would grant OpenAI a 10 percent stake in the chipmaker, potentially providing a substantial capital boost for its ambitious data center expansion plans in the years ahead.
This collaboration underscores a broader trend in the tech industry, where companies are pouring hundreds of billions of dollars into constructing new data centers. Giants like OpenAI, Amazon, Google, Meta, and Microsoft are collectively projected to spend over $325 billion on these facilities by the close of this year alone.
Established tech behemoths such as Amazon, Microsoft, and Google, with their immense annual profits, can typically self-finance such massive infrastructure projects. However, newer and rapidly expanding entities like OpenAI often find themselves needing to secure significant external funding or incur substantial debt to build out their computing capabilities.
As part of its “Stargate Project,” OpenAI had previously indicated plans to partner with cloud provider Oracle and Japanese conglomerate SoftBank to invest over $400 billion in new data centers across the United States. Recognizing the vast financial requirements for these projects, OpenAI has been actively exploring innovative funding solutions to bridge this considerable gap.
The recent deal with Nvidia exemplified this strategy: in addition to committing to deploy Nvidia’s chips, OpenAI also received a $100 billion investment from the chipmaker. This includes an initial $10 billion, with an additional $90 billion slated over several years, highlighting OpenAI’s approach of securing investments from the very companies whose products and services are crucial to its operations.
(It’s worth noting that The New York Times initiated a lawsuit against OpenAI and Microsoft in 2023, alleging copyright infringement related to AI systems, claims which both companies have denied.)