OpenAI has officially entered into an agreement with Oracle to procure computing infrastructure worth $300 billion over a five-year period, forming a key component of its previously announced Project Stargate data center initiative. This deal ranks among the largest cloud contracts in history, representing a massive investment that far exceeds OpenAI's current revenue levels. It underscores the rapid pace at which Silicon Valley firms are pouring resources into AI data center infrastructure. According to reports, the collaboration between OpenAI and Oracle will deliver 4.5 gigawatts of data center computing capacity—approximately equivalent to the electricity generated by two Hoover Dams or sufficient to power around four million U.S. households. The announcement follows Oracle’s recent stock surge of over 42% on Wednesday, driven by its report of $317 billion in future contract revenues for the latest quarter. During a conference call, Oracle CEO Safra Catz revealed that her company had signed agreements with three clients, one of which is now confirmed as OpenAI. Initial indications of the deal emerged in June when Oracle disclosed in a regulatory filing that it had agreed to a cloud services contract expected to generate more than $30 billion in revenue by the 2027 fiscal year, with increasing returns as more infrastructure comes online. A month later, OpenAI revealed plans to acquire 4.5 gigawatts of computing capacity from Oracle but did not disclose the full contract scope at the time. Project Stargate was first unveiled in January at the White House, where OpenAI CEO Sam Altman and Oracle founder and CTO Larry Ellison joined U.S. President Donald Trump. Trump emphasized that the initiative aims to solidify U.S. leadership in artificial intelligence. Initially, the partners pledged to build data centers worth at least $100 billion, with plans to expand this effort to $500 billion in the coming years. OpenAI later confirmed that Stargate serves as the brand for all its data center infrastructure initiatives, which also involve strategic investor SoftBank Group Corp. Following the Stargate announcement, OpenAI has already secured more than half of its $500 billion target, with construction underway in Abilene, Texas. OpenAI and Oracle are not alone in their rush to build AI data centers. Rivals including Amazon Web Services, Microsoft, Google LLC, and Meta Platforms have collectively committed $300 billion this year alone to develop their own large-scale data center networks. However, the OpenAI-Oracle partnership may be seen by some as carrying higher risk. OpenAI remains far from profitability and is burning through significant amounts of capital. In June, it reported annual revenues of around $1 billion—only a fraction of the $60 billion it spends annually on data center construction and rented computing power. Oracle, in turn, is placing a substantial portion of its future revenue on a single client. The company may need to take on debt to fund the purchase of AI chips required for its planned data centers. Altman has already committed to several high-risk ventures, including a reported partnership with Broadcom to develop custom AI chips and an effort to build a smartphone to rival Apple's iPhone. OpenAI, which has raised billions in funding, is burning through cash at a rate unmatched by most startups. Altman reportedly told investors last year that he expects OpenAI to reach profitability no sooner than 2029 and that it may require up to $44 billion to get there. In essence, the OpenAI-Oracle deal represents a bold bet that ChatGPT’s explosive growth will continue accelerating and will ultimately be adopted by billions of global consumers and thousands of enterprises. However, the venture faces stiff competition from rivals like Google and other AI startups such as Anthropic PBC, along with growing tensions with Microsoft, its primary financial backer. According to reports, Oracle carries a heavier debt burden relative to its cash reserves compared to major competitors like Microsoft, Amazon, and Meta. Its investments in AI infrastructure already exceed its cash flow, with a total debt-to-equity ratio of 427%, compared to Microsoft’s 32%. Despite the risks, Oracle still has options if things don’t go as planned, noted Holger Mueller of Constellation Research Inc. He added that Oracle’s recent stock surge, the biggest single-day jump since 1992, gives the company some flexibility to take on risk. “Oracle may have entered the cloud race later than its hyperscale peers, but that isn’t necessarily a disadvantage, as its recent data center developments mean it now possesses the most modern cloud infrastructure,” the analyst explained. “Even better, this infrastructure is optimized for Oracle databases, which has turned out to be ideal for AI workloads. It has also made early and substantial investments in Nvidia GPUs and has ample available capacity.” In short, Oracle offers everything OpenAI needs, which is why it was chosen ahead of other cloud infrastructure providers, Mueller said. The biggest short-term risk lies in potential obstacles during Oracle’s data center construction phase, while the mid-term concern is whether OpenAI can achieve sufficient growth to justify its $60 billion annual investment in AI computing. “However, Larry Ellison and Safra Catz are betting that even if OpenAI doesn’t succeed, they can repurpose this capacity elsewhere—for example, by offering AI training infrastructure to Oracle database clients,” Mueller added. “If needed, they do have promising alternative revenue options.” Additionally, OpenAI and Oracle have committed to a separate initiative to construct a major data center complex in the United Arab Emirates, part of an agreement negotiated by Trump and the UAE government. This venture is a joint effort involving SoftBank, UAE investment firm G42, and other stakeholders.