Brookfield and Bloom Energy Expand $25B Partnership to Finance On-Site AI Power as Grid Delays Mount
July 1, 2026
Brookfield and Bloom Energy Expand $25B Partnership to Finance On-Site AI Power as Grid Delays Mount
Brookfield Asset Management announced Tuesday it has expanded its financing partnership with Bloom Energy to $25 billion, up from an initial $5 billion commitment, in a move aimed at accelerating on-site power generation for hyperscalers and AI developers facing prolonged grid interconnection delays. The fivefold increase signals a broader strategic shift in which investors are treating energy certainty not as a routine utility expense, but as a core, financeable component of AI infrastructure.
The partnership, first established in October 2025, was originally designed to fund Bloom fuel-cell projects. With the expanded framework, the companies plan to fund projects globally and compress deployment timelines by bundling capital with on-site power from day one, allowing data center projects to advance even while awaiting utility connections. Brookfield’s approach integrates capital and generation from the outset, enabling operators to align investments across power, compute, and data center infrastructure.
Neil Osnato, founder of Persistence Analytics Group, said the expansion represents more than just a larger financing commitment, but cautioned against treating behind-the-meter generation as a standalone asset class. “I don’t think behind-the-meter power should automatically be viewed as a new asset class in isolation,” Osnato said. “Rather, it is part of a broader transition where energy certainty becomes a financeable asset. Investors are increasingly allocating capital not just to servers and buildings, but to the ability to deliver dependable megawatts on schedule when the grid cannot.”
AI infrastructure priorities have shifted rapidly. What began as a race for GPUs now hinges on securing dependable megawatts on schedule. Rather than funding individual fuel-cell installations, Brookfield and Bloom said the framework is designed to compress deployment timelines by bundling capital with power, compute, and data center infrastructure from the start. “Together, the companies continue to advance a new model for AI factories that integrates power, compute, data center infrastructure, and capital from the outset,” they said in a joint statement.
Brookfield said the expansion fits within its AI Infrastructure Fund, launched in late 2025 with a goal of deploying $100 billion across AI factories, power solutions, compute infrastructure, and strategic capital partnerships. The firm has already invested more than $100 billion in digital infrastructure and clean power assets.
The Brookfield–Bloom partnership enters a competitive market for AI power infrastructure. Companies across the sector are pursuing the same opportunity through different technologies. GE Vernova’s gas turbine approach for AI campuses and Wärtsilä’s 790 MW off-grid Texas data center project reflect a broader push to bring dedicated generation closer to large AI loads. Brookfield and Bloom aim to differentiate by embedding financing into the on-site generation model, making capital itself a lever for schedule assurance rather than merely a means of purchasing equipment.
Osnato noted that the next phase will be less about proving on-site generation works than about validating the assumptions behind these investments. Investors will increasingly scrutinize whether projected AI loads materialize, whether fuel supplies and operating costs remain sustainable, how on-site systems interact with utility planning, and who ultimately bears stranded-asset risk if demand shifts. For operators, access to capital to fund on-site generation may prove as critical as access to the generation technology itself. As utilities contend with longer interconnection queues and rising large-load requests, developers who can secure both power and financing could gain an advantage in bringing new AI capacity online.
Whether this fivefold expansion is an early indicator or the beginning of a broader investment model remains to be seen. But as AI developers compete for dependable megawatts as aggressively as they compete for GPUs, energy certainty is emerging as infrastructure in its own right for AI development.
Source: datacenterknowledge
