Oracle’s AI infrastructure expansion is being financed through several channels: company debt and equity, customer-supplied hardware, and multibillion-dollar project loans tied to new data centers. That funding mix is helping Oracle pursue record cloud demand, while enterprise customers evaluating future capacity still need clarity on construction schedules, utility approvals, and contract terms.
The company ended fiscal 2026 with $638 billion in remaining performance obligations, or contracted revenue expected to be recognized later. Oracle also reported negative $23.7 billion in free cash flow after heavy cloud infrastructure investment and said it expects to raise about $40 billion through debt and equity financing in fiscal 2027.
How Oracle is funding the buildout
Oracle’s fiscal 2026 results show the scale of its financing plan. It raised $43 billion in debt and $5 billion in equity during the year. The spending push has also accompanied a broader restructuring around AI and cloud infrastructure.
Prepaid or customer-supplied hardware tied to large AI contracts totaled $75 billion, Oracle said, reducing the amount it must spend on GPUs. That support does not cover every development expense. Land, construction, utility connections, and AI-era power and cooling requirements still add specialized engineering and equipment costs.
A $38 billion project-financing package for Vantage Data Centers campuses in Texas and Wisconsin shows how some Oracle-linked capacity is being funded through developer-level loans. JPMorgan Chase and Mitsubishi UFJ Financial Group led the financing for facilities expected to be leased to Oracle and used for OpenAI workloads.
By April 2026, lenders had syndicated nearly all of the package, with less than $1 billion reportedly still to be sold. The Texas campus is expected to include 10 data centers, with the first building targeted for the second half of 2026 and the rest by the end of 2028.
Wisconsin presents a separate utility issue. In July 2026, the Financial Times reported that Oracle could be required to provide more than $7 billion in collateral under rules intended to shield electricity customers from costs associated with very large power users. Oracle is challenging the requirement.
What shapes OCI delivery timelines
Organizations buying capacity tied to a new or expanding facility should treat Oracle’s backlog as evidence of demand, not confirmation that a specific deployment is already operating and ready.
Four contract checks deserve attention:
- Capacity status: Confirm whether the resources are operating, under construction, or dependent on a future facility.
- Delivery date: Require a contractual availability date instead of relying on a developer’s estimated schedule.
- Delay remedies: Define substitution rights, service credits, termination rights, and alternative-region options.
- Infrastructure dependencies: Ask whether utility approvals, grid connections, unfinished construction, or AI networking bottlenecks could affect usable capacity.
Contracts should define the capacity, delivery date, and remedies for delays. Oracle’s global backlog and infrastructure spending provide scale, but customer agreements still determine delivery dates and protections.
Fiscal 2027 disclosures, the Wisconsin utility dispute, and major campus milestones will show how quickly Oracle’s financing strategy is producing capacity customers can actually use.
Read more: Large infrastructure agreements can leave customers with unanswered questions about availability, pricing, and control, as shown by the Google-SpaceX $30 billion compute deal.