AMD said Monday that it has reached a deal to supply OpenAI with hundreds of thousands of AI chips and offered the firm a potential 10% stake in the company.
Shares in AMD soared 27% Monday morning, adding nearly $80 billion to the firm’s market cap as investors cheered the firm’s addition as one of OpenAI’s key computing partners.
The deployment agreement covers a whopping six gigawatts – enough to power 5 million US households – over several years beginning in the second half of 2026.
AMD said the deal is expected to bring in tens of billions of dollars in revenue while accelerating OpenAI’s push into AI infrastructure.
“This partnership is a major step in building the compute capacity needed to realize AI’s full potential,” OpenAI CEO Sam Altman said in a statement Monday.
“AMD’s leadership in high-performance chips will enable us to accelerate progress and bring the benefits of advanced AI to everyone faster.”
OpenAI plans to build a one-gigawatt facility based on upcoming AMD Instinct MI450 chips starting next year.
As part of the deal, AMD has issued OpenAI a warrant for up to 160 million shares of common stock, structured to vest as specific milestones are achieved, like the initial 1 gigawatt deployment.
If the ChatGPT owner exercises the full warrant, it will receive a roughly 10% stake in the chipmaker.
“We are thrilled to partner with OpenAI to deliver AI compute at massive scale,” AMD CEO Lisa Su said in a statement.
“This partnership brings the best of AMD and OpenAI together to create a true win-win enabling the world’s most ambitious AI buildout and advancing the entire AI ecosystem.”
It’s one of the largest GPU deployment agreements yet – and comes just two weeks after OpenAI unveiled a $100 billion chips deal with rival Nvidia.
The new deal with AMD will help reduce OpenAI’s reliance on a single vendor, putting a dent in Nvidia’s industry-wide dominance. Shares in Jensen Huang’s Nvidia dipped 1.1% Monday.
OpenAI is also in talks with Broadcom to build custom chips for its next generation of GPUs.
The deals have resulted in a broader, more complex supply chain, reminiscent of a larger trend across the AI sector.
Nvidia is supplying the capital needed to buy chips while Oracle is helping build massive data centers to generate the required power.
AMD and Broadcom are supplying the GPUs, while OpenAI is bringing in the demand.
It’s similar to the Stargate project, a $500 billion venture to quickly build AI infrastructure in the US involving OpenAI, Oracle and SoftBank that President Trump announced in January.
OpenAI’s first data center in the project is already up-and-running in Abilene, Texas, and using Nvidia chips. Construction is ongoing at the site to build up capacity.
Planned data centers in New Mexico, Ohio and the Midwest are expected to rely on a mix of suppliers, including AMD.
OpenAI also announced last month that it had signed an agreement to restructure OpenAI into a for-profit entity – despite legal action from Altman’s rival, Elon Musk, who has attempted to thwart the restructuring.
