DOE Announces More Than $100 Billion Paducah AI Campus Partnership
The U.S. Department of Energy on July 29 announced a partnership with Brookfield, NextEra Energy and regional utilities to redevelop part of its Paducah, Kentucky site as a 1.8 GW AI and high-performance computing campus. DOE described the more than $100 billion project as privately funded; NextEra intends to add 2 GW of gas generation and up to 2.6 GW of battery storage.
The U.S. Department of Energy on July 29 announced a partnership with Brookfield, NextEra Energy, Big Rivers Electric Power Corporation, Jackson Purchase Energy Cooperative and Paducah Power System to redevelop part of the DOE Paducah Site in western Kentucky as an AI and high-performance computing campus. DOE described the project as a more than $100 billion privately funded investment expected to create about 8,000 construction jobs and 600 permanent jobs.
DOE describes a 1.8 GW AI and high-performance computing campus and expects construction to be completed in 2031. Data Center Knowledge reports that the partners are targeting more than 1.2 GW of compute capacity and up to 1.8 GW of utility-delivered capacity by 2032, with Brookfield developing and operating the campus.
Dedicated power and storage
DOE said NextEra Energy intends to develop 2 GW of new grid-connected natural gas-fired generation, upgrade transmission infrastructure and deploy up to 2.6 GW of battery energy storage at or near the campus. The department said generation and storage would exceed the campus's needs, allowing excess electricity to serve the regional grid.
The power service agreement still requires approval from the Kentucky Public Service Commission. Data Center Knowledge also cautions that gas-generation and battery nameplate ratings should not be treated as equivalent firm capacity; delivery will depend on transmission, fuel supply, battery operating assumptions and grid conditions.
For AI infrastructure operators, the proposal shows how electricity procurement, transmission access and storage are becoming part of campus design rather than an external utility assumption. Dedicated resources may reduce dependence on existing local generation, but construction still depends on interconnection, permitting and regulatory execution.
A former enrichment site
Associated Press reporting says the Paducah Gaseous Diffusion Plant was used for uranium enrichment connected to U.S. weapons and commercial nuclear power before closing in 2013. DOE continues a long-running cleanup of the roughly 3,550-acre site.
DOE selected Brookfield after a November 2025 request for offers to lease land and develop an AI data center campus at Paducah. It selected NextEra to build and own the dedicated energy infrastructure, while regional utilities would provide wholesale and retail service and act as a community partner.
Data Center Knowledge reports that definitive agreements still require negotiation and execution. Together with regulatory approvals, interconnection and permitting, those conditions mean the headline investment and capacity targets remain a development plan rather than completed infrastructure.
Key Points
- 1DOE's Paducah partnership combines a proposed 1.8 GW AI campus with 2 GW of gas generation and up to 2.6 GW of battery storage.
- 2The more than $100 billion plan illustrates how hyperscale AI campuses increasingly pair compute capacity with dedicated power and transmission infrastructure.
- 3Definitive agreements, regulatory approvals, interconnection and permitting remain key execution milestones.
Scoring Rationale
The proposed investment and gigawatt-scale compute capacity make this a major AI infrastructure development, especially because it pairs data-center construction with dedicated generation and storage. The project remains subject to definitive agreements and regulatory approvals, so its practical impact depends on execution, permitting and grid development.
Sources
Primary source and supporting public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
