OpenAI CEO Sam Altman has long stated that his company’s biggest bottleneck is not ideas or talent, but raw computing power. With the announcement of a multibillion-dollar partnership with AMD, Altman has taken a giant leap toward removing that constraint, securing the massive infrastructure needed to power OpenAI’s ambitious goals.
The agreement provides OpenAI with a dedicated supply of hundreds of thousands of AMD’s cutting-edge AI chips. This vast arsenal of GPUs is the lifeblood of AI development, enabling the training of ever-larger and more capable models. The deal is structured for the long term, with chip deployments beginning in the second half of 2026 and continuing for several years.
In a statement, Altman confirmed that the deal with AMD will help OpenAI build sufficient AI infrastructure to meet its needs. The sheer scale of the project is breathtaking, involving six gigawatts of power—a clear signal of the computational intensity required to stay at the forefront of the AI race. As part of the plan, OpenAI will begin building a one-gigawatt facility next year using AMD’s next-generation MI450 chips.
This strategic alliance is more than a simple transaction. OpenAI has been given the option to purchase a 10% stake in AMD through a warrant for 160 million shares. This move aligns the incentives of both companies, ensuring they work in close concert to push the boundaries of AI technology. It follows another major move by OpenAI to secure its supply chain, coming just a week after Nvidia announced a planned $100 billion investment.
The financial implications are profound. AMD’s market capitalization skyrocketed by $80 billion post-announcement, and the company now projects over $100 billion in new revenue over four years, driven by this anchor partnership. For OpenAI, while the deal’s funding remains undisclosed, it is a clear declaration of its intent to spare no expense in the quest for artificial general intelligence.