Core Scientific Secures $14B AMD Deal, Doubles AI Capacity to 1.1 GW

Core Scientific Expands AI Infrastructure Through Major AMD Partnership

Core Scientific has secured a 15-year infrastructure agreement with AMD that doubles the company's leased AI data center capacity to approximately 1.1 GW. The deal covers 530 MW across five U.S. campuses and represents more than $14 billion in potential base contracted revenue, according to details first reported by Data Center Knowledge.

The agreement, announced July 28 alongside Core Scientific's second-quarter results, begins deployment in 2027 and grants AMD exclusive rights to reserve up to an additional 2 GW of future capacity. The deployment will use Core Scientific facilities to host infrastructure built around AMD Instinct GPUs, EPYC processors, and the ROCm software platform.

Why it matters

This deal illustrates how AI infrastructure competition has evolved beyond chip performance to encompass the entire deployment stack. By securing long-term data center capacity, AMD is addressing one of the most significant bottlenecks in AI deployment: access to power-ready, liquid-cooled facilities capable of supporting high-density compute clusters. The agreement also demonstrates Core Scientific's successful pivot from cryptocurrency mining to AI infrastructure, a transition that has driven colocation revenue from $10.6 million to $136.7 million year-over-year.

Infrastructure Economics and Execution Timeline

Core Scientific disclosed that fully delivered AI data center capacity costs approximately $11 million to $12 million per megawatt, including construction, electrical infrastructure, cooling systems, commissioning, and utility integration. At that rate, the initial AMD deployment represents roughly $6 billion in infrastructure investment.

Construction is underway at the company's Pecos, Texas campus, with deliveries scheduled to begin in the first half of 2027. Development is also progressing at campuses in Hunt, Texas; Muskogee, Oklahoma; Auburn, Alabama; and Dalton, Georgia, with deployments continuing through 2028.

Mathew Hein, AMD's chief strategy officer for corporate development, said the agreement "expands access to the infrastructure our customers need to deploy AMD AI solutions at scale."

Analyst Perspective on AMD's Strategy

Matt Kimball, vice president and principal analyst for data center technologies at Moor Insights & Strategy, characterized the deal as AMD using its balance sheet to remove deployment barriers rather than simply becoming a data center landlord.

"It's not about selling accelerators at this point, it's about selling a deployment path," Kimball told Data Center Knowledge. He cautioned that the $14 billion figure represents base contracted revenue over 15 years rather than a single booking, while the additional 2 GW reflects reservation rights rather than firm customer commitments.

From Bitcoin to AI: A Business Transformation

The expansion underscores Core Scientific's rapid business transformation. Colocation revenue climbed to $136.7 million in the second quarter from $10.6 million a year earlier, while digital asset self-mining revenue fell to $21.5 million as the company shifted power from bitcoin mining to AI workloads.

Core Scientific now has two AI customers, each committed to more than 500 MW across five campuses. As of mid-July, the company was billing customers for 437 MW of capacity, equivalent to an annualized colocation revenue run rate of approximately $635 million.

CEO Adam Sullivan said Core Scientific has identified more than 2 GW of additional development opportunities that could begin coming online between late 2028 and 2030.

These details were first reported by Shane Snider at Data Center Knowledge.