OpenAI Teams With AMD to Develop AI Data Centers

OpenAI launched a multibillion-dollar partnership with chip designer Advanced Micro Devices (AMD).

The collaboration will see the companies develop artificial intelligence data centers running on AMD processors, according to a Monday (Oct. 6) press release.

“This partnership is a major step in building the compute capacity needed to realize AI’s full potential,” OpenAI co-founder and CEO Sam Altman said in the release. “AMD’s leadership in high-performance chips will enable us to accelerate progress and bring the benefits of advanced AI to everyone faster.”

With this deal, OpenAI pledged to purchase 6 gigawatts worth of AMD’s chips, beginning with the MI450 chip next year, the release said.

It will buy the chips either directly or through OpenAI’s cloud computing partners, The Wall Street Journal (WSJ) reported Monday.

As part of the agreement, AMD issued OpenAI a warrant for up to 160 million shares of AMD common stock, designed to vest as specific milestones are reached, according to the release. The first tranche vests with the initial 1-gigawatt MI450 chip deployment, with more tranches vesting as purchases scale up to 6 gigawatts.

“Vesting is further tied to AMD achieving certain share-price targets and to OpenAI achieving the technical and commercial milestones required to enable AMD deployments at scale,” the release said.

The deal comes on the heels of chipmaker Nvidia’s $100 billion pledge to OpenAI, a deal that, when realized, would be the largest private company investment on record.

AMD CEO Lisa Su said the deal will result in tens of billions of dollars in new revenue for the chip company over the next half-decade, per the WSJ report. She was interviewed jointly with Altman, who said demand for inference computing has jumped in tandem with the rise of AI tools like large language models.

“It’s hard to overstate how difficult it’s become” to get sufficient computing power, Altman said, according to the report. “We want it super fast, but it takes some time.”

Inference is when a trained model is applied to new inputs in real time, a “constant, real-time activity” that never stops, PYMNTS reported Sept. 22. Keeping ChatGPT online alone reportedly costs OpenAI tens of millions of dollars each month.

Altman and Su said their collaboration gives both companies incentives to commit to the boom in AI infrastructure, WSJ reported.

“It’s a win for both of our companies, and I’m glad that OpenAI’s incentives are tied to AMD’s success and vice-versa,” Su said, per the report.

For all PYMNTS AI coverage, subscribe to the daily AI Newsletter.

The post OpenAI Teams With AMD to Develop AI Data Centers appeared first on PYMNTS.com.