San Francisco, October 2025 – In a move that could reshape the future of the artificial intelligence (AI) chip market, OpenAI and Advanced Micro Devices (AMD) have announced a multi-year agreement that positions AMD as a key strategic partner in OpenAI’s massive infrastructure expansion.
The deal, valued in the billions, could give OpenAI up to a 10% stake in AMD if certain milestones are achieved, marking one of the largest GPU deployment commitments in AI history. Following the announcement, AMD’s stock skyrocketed 23.71% on Monday, while rival Nvidia’s shares dipped by 1%.
A 6-Gigawatt GPU Deployment Plan
As part of the partnership, OpenAI will deploy 6 gigawatts of AMD’s Instinct GPUs across several years and multiple hardware generations. The rollout will begin in the second half of 2026 with an initial 1-gigawatt deployment.
Greg Brockman, President of OpenAI says:
“This is so core to our mission if we really want to scale to reach all of humanity. We’re already unable to launch many features in ChatGPT and other products because of limited compute power. This partnership will change that.”
A Unique Equity Structure
To cement the deal, AMD issued OpenAI a warrant for up to 160 million shares of AMD common stock. These shares will vest in stages as OpenAI scales its GPU usage from 1 gigawatt to 6 gigawatts, alongside AMD meeting share price and technical milestones.
If OpenAI exercises the full warrant, it could acquire about 10% ownership in AMD, based on today’s outstanding shares — making this a landmark equity-and-compute partnership in the AI sector.
A Challenge to Nvidia’s Dominance
For years, Nvidia has dominated the AI accelerator market, but this deal gives AMD a powerful entry into the generative AI boom. AMD CEO Lisa Su said the partnership will accelerate innovation across the ecosystem:
“AI is on a 10-year growth path, and foundational compute is essential. Partnerships like this bring the ecosystem together to deliver the best technologies. We’re excited about the opportunities this creates.”
The agreement also diversifies OpenAI’s supplier base, reducing its dependence on Nvidia. Just two weeks ago, OpenAI unveiled a $100 billion deal with Nvidia, which included equity and long-term GPU supply.
A $1 Trillion AI Infrastructure Bet
Combined, OpenAI’s recent Nvidia and AMD agreements represent a commitment of nearly $1 trillion in AI infrastructure buildout. Each gigawatt of compute capacity is estimated to cost around $50 billion in construction.
These deals account for a significant portion of OpenAI’s ambitious 23-gigawatt roadmap, which spans data centers in Texas, New Mexico, Ohio, and the Midwest. The initiative, internally known as the Stargate project, positions OpenAI not just as a software leader but also as one of the largest infrastructure builders in AI.
The Bigger Picture: A Circular AI Economy
The OpenAI-AMD deal highlights the increasingly interconnected nature of the AI industry. Nvidia is funding chips to fuel its own dominance, Oracle is building data centers, Broadcom is in talks for custom chip design, and AMD is now supplying large-scale GPU deployments.
OpenAI sits at the center, anchoring demand across this tightly wound supply chain. Analysts, however, warn that such an interdependent structure could face risks if any single link falters.
What This Means for the Future
For AMD, the deal is a long-awaited breakthrough, validating its Instinct GPU roadmap and securing a flagship customer in the AI race. For OpenAI, it ensures the compute power needed to keep advancing ChatGPT and future AI models.
Su summed it up best:
“This creates a true win-win, enabling the world’s most ambitious AI buildout while advancing the entire AI ecosystem.”
With two of the world’s largest chipmakers now directly tied to its growth, OpenAI has solidified its role as both a software pioneer and an infrastructure powerhouse — setting the stage for the next explosive phase of the global AI boom.
Read more similar articles at OpenAI Concludes DevDay 2025 with the Successful Launch of ‘Agent Builder’