Oracle Project Jupiter's delay raises new financing and execution risks for scheduled AI data center buildout.

The delay affecting Oracle's Project Jupiter data-center campus in New Mexico is creating consequences beyond the construction schedule, prompting lenders and investors to examine how the rapidly expanding AI infrastructure market is financed.

Project Jupiter is being developed by STACK Infrastructure, a unit of Blue Owl, to support OpenAI. Oracle recently issued a force majeure notice related to the project, citing delays that could affect its ability to meet contractual obligations. According to people familiar with the project and its financing, completion is now expected to be delayed by about a year.

Blue Owl has approximately $3 billion of equity invested in the project, making the delay economically significant. The investment generates lower returns during construction, with higher returns expected once the data center becomes operational. The delay thus defers when those higher returns begin.

The immediate issue surrounding Project Jupiter is access to power. Large AI data centers require enormous amounts of electricity, and securing generation and grid connections has become one of the most difficult aspects of developing new campuses. Oracle invoked force majeure in connection with delays securing power for the site, according to a person familiar with the matter. Blue Owl has maintained that the notice does not change the companies' financial commitments to the multiyear project.

However, the situation has drawn attention from financial institutions, as AI data centers are increasingly financed through complex combinations of corporate commitments, project-level capital, and debt. For lenders, delays can alter when projects begin generating revenue while construction costs and financing expenses continue to accumulate.

Project Jupiter is now being examined as an example of how contracts can distribute risk when an AI data-center project encounters delays outside the immediate control of its developers. Force majeure provisions allow companies to seek contractual protection when circumstances beyond their control make it difficult to fulfill their obligations. Lawyers involved in the sector say these provisions are becoming more common in data-center development agreements as projects become larger and more complex. That is changing how lenders evaluate new AI infrastructure deals.

Instead of focusing primarily on the creditworthiness of the technology company supporting a project, investors are increasingly examining questions around construction schedules, power availability, contractual obligations, and who ultimately bears the financial consequences of delays. The Project Jupiter situation therefore provides a real-world example of how those provisions could affect the economics of large AI campuses.

These financing concerns arrive as technology companies dramatically increase spending on AI infrastructure. Reuters reported that AI-related capital spending by the six largest U.S. technology companies is projected to reach approximately $1 trillion in 2027, according to Moody's. That scale of investment creates another challenge: financial markets must absorb enormous volumes of debt and other external financing while investors assess the long-term economics of AI infrastructure. Morgan Stanley estimates that AI infrastructure could require approximately $1.5 trillion in external financing by 2028. That capital will be needed across data centers, power generation, networking equipment, and other infrastructure supporting increasingly compute-intensive AI workloads. The larger the capital requirement becomes, the more closely investors will scrutinize individual projects and the contractual structures supporting them.

Project Jupiter is not an isolated example of challenges affecting large AI infrastructure developments. Reuters reported that other proposed projects are also encountering financing, permitting, and community-related obstacles. SB Energy, which is developing an Ohio campus intended to serve OpenAI, delayed its planned initial public offering during the same week. Blue Owl also previously faced difficulties syndicating financing for a roughly $4 billion data-center project in Lancaster, Pennsylvania, anchored by CoreWeave. Elsewhere, projects involving companies such as Meta and Related Digital have faced opposition from local communities, while QTS terminated its planned Digital Gateway data-center project in Virginia after years of planning and regulatory review.

These developments illustrate the growing complexity of building AI infrastructure at scale. Access to capital is only one component. Developers must also secure power, obtain permits, address local concerns, and construct facilities within timelines that support their customers' computing requirements.

The financing concerns do not necessarily indicate weaker demand for AI computing. Instead, they point to a widening gap between demand for computing capacity and the practical requirements for delivering it. AI companies and cloud providers can commit billions of dollars to future computing capacity, but those commitments still depend on physical infrastructure being completed on schedule. That makes power availability, construction timelines, and contractual risk increasingly important to investors.

For infrastructure developers and their financial partners, the Project Jupiter delay could therefore serve as a reference point when structuring future AI data-center deals. As the industry moves toward larger campuses and increasingly capital-intensive projects, financing agreements will have to account for risks that were less prominent when data centers were smaller and power requirements were lower. The next phase of the AI infrastructure buildout will depend not only on demand for computing, but also on whether developers, utilities, technology companies, and financial institutions can structure projects that remain viable when construction and power timelines change.