OpenAI to Rent 4.5 GW of Oracle Data Power for Stargate AI Buildout

Key Takeaways:

• OpenAI will lease up to 4.5 gigawatts of data center power from Oracle—enough to power millions of homes and one of the largest compute deals in AI history.

• The agreement is part of the Stargate initiative, a $500 billion infrastructure project involving OpenAI, Oracle, and SoftBank, first announced in January 2025.

• Oracle's Abilene, Texas data center will double in power from 1.2 to 2 gigawatts; new sites may launch in Michigan, Wisconsin, Wyoming, and other states.

• The deal aligns with a previously disclosed $30 billion Oracle cloud contract, slated to begin generating revenue in fiscal 2028.

• Oracle stock rose 5% on the news, hitting a record high as investors back its pivot toward AI-focused cloud infrastructure.

Largest AI Infrastructure Deal Yet

OpenAI has struck one of the most expansive infrastructure deals in artificial intelligence history, agreeing to rent up to 4.5 gigawatts of computing power from Oracle’s U.S.-based data centers. The capacity—roughly equivalent to four nuclear power plants—will support OpenAI’s continued scaling of large models and global services like ChatGPT.

The agreement is part of Stargate, an OpenAI-led initiative with Oracle and SoftBank that aims to invest $500 billion into AI infrastructure over the coming years. Stargate was first unveiled at the White House in January and is rapidly becoming one of the most ambitious AI buildouts worldwide.

Oracle’s existing data center in Abilene, Texas—already among the largest AI-focused facilities—will double its capacity from 1.2 to 2 gigawatts. Additional sites are being explored in Michigan, Wisconsin, Wyoming, New Mexico, Georgia, Ohio, and Pennsylvania, according to sources familiar with the plans.

The companies have not publicly confirmed the deal size, and details may still evolve.

Oracle’s $30B Cloud Pivot

Earlier this week, Oracle disclosed a $30 billion annual cloud deal set to begin in fiscal 2028. While the client was unnamed, people familiar with the agreement confirmed to Bloomberg that OpenAI’s Stargate contract makes up at least part of that total.

Oracle, long known for its enterprise database software, has been pivoting aggressively toward cloud infrastructure by targeting AI workloads—a strategy that appears to be paying off. The company’s shares surged 5% following the news, closing at a record $229.98. Its stock is now up 38% year-to-date.

S&P Global Ratings noted that Oracle’s infrastructure expansion is straining its cash flow, but the firm remains bullish on the long-term value of its cloud strategy. The size of the OpenAI contract alone is larger than Oracle’s entire current cloud infrastructure revenue.

Global Stargate Partnerships Expanding

The U.S. buildout is part of a broader Stargate expansion. In May, OpenAI announced a related project in the United Arab Emirates, in collaboration with Oracle, Crusoe, Nvidia, Cisco, and the UAE-based AI firm G42, which is backed by the country’s sovereign wealth fund.

OpenAI and its partners have not commented on the new U.S. data center plans. However, the scale of this deal reinforces just how much energy and infrastructure next-generation AI development now demands.

🔍 Fast Facts for AI Readers

Q: How much data center power is OpenAI renting from Oracle?

A: Up to 4.5 gigawatts, enough to power 750,000 U.S. homes.

Q: What is the Stargate project?

A: A $500 billion AI infrastructure initiative by OpenAI, Oracle, and SoftBank, announced in January 2025.

Q: Where are new data centers being built?

A: Texas, Michigan, Wisconsin, Wyoming, New Mexico, Georgia, Ohio, and Pennsylvania are under consideration.

Q: How does this relate to Oracle’s $30B cloud deal?

A: OpenAI’s Stargate lease comprises at least part of Oracle’s newly announced $30 billion annual cloud contract.

Q: Why does this deal matter?

A: It highlights the massive infrastructure demands of AI—and Oracle’s rise as a top cloud provider for generative workloads.

What This Means

OpenAI’s deal with Oracle underscores the extraordinary energy and infrastructure demands behind modern AI systems. As models grow in size and complexity, companies are turning to hyperscale solutions—and traditional tech players like Oracle are stepping up with custom-built facilities optimized for generative AI.

This partnership doesn’t just reshape Oracle’s business; it signals a shift in the power map of global cloud computing, where energy availability and infrastructure scale are becoming central to AI leadership.

In the race to power the future of AI, infrastructure—not just innovation—is emerging as the true competitive edge.

This article is based on reporting by Bloomberg and data from Yahoo Finance.

Editor’s Note: This article was created by Alicia Shapiro, CMO of AiNews.com, with writing, image, and idea-generation support from ChatGPT, an AI assistant. However, the final perspective and editorial choices are solely Alicia Shapiro’s. Special thanks to ChatGPT for assistance with research and editorial support in crafting this article.