Bloom Energy Corporation and Brookfield Corporation have announced a $5 billion (approx. AED18.37bn) strategic alliance to deploy infrastructure dedicated to artificial intelligence (AI) worldwide. The partnership will integrate Bloom Energy’s fuel cell technology into upcoming “AI factories”, a concept combining high-performance computing, dedicated power supply, and data centre architecture. It marks Brookfield’s first investment in its AI infrastructure strategy.
A response to traditional grid constraints
With rising demand for computing power, on-site energy production capacity is becoming a central factor in competitiveness. Bloom Energy’s fuel cells are designed to deliver continuous and scalable power without relying on existing electricity grids. Brookfield plans to deploy this technology at several locations, including a European site to be announced before year-end.
The rise of generative AI models is intensifying pressure on global energy systems. Industry data estimates that AI-related data centres in the United States could exceed 100 gigawatts of power demand by 2035. Integrating off-grid solutions has thus become a strategic necessity in an increasingly constrained energy market.
Brookfield’s first concrete step in its AI strategy
The partnership launches Brookfield’s AI-focused investment strategy, covering specialised factories, computing infrastructure and energy partnerships. This new pillar complements Brookfield’s existing digital portfolio, which exceeds $100 billion globally across telecom, energy and digital services.
Brookfield, managing over $550 billion in critical assets, recently completed major U.S. deals in energy infrastructure with Compass Datacenters, Duke Energy Florida and Colonial Enterprises. It also signed a 3-gigawatt hydroelectric supply agreement with Google.
Bloom Energy, for its part, has already deployed several hundred megawatts of capacity with clients such as American Electric Power (AEP), Equinix and Oracle. Its on-site power generation model enhances the energy autonomy of data centres while accelerating deployment in constrained environments.
Redesigning the data centre model
The two companies’ collaboration is based on an integrated industrial approach. The “Lean AI factory” concept is built on a joint design from day one of power, data architecture, and computing needs. This strategy aims to accelerate deployment while reducing dependency on overstretched public infrastructure.
“Behind-the-meter power solutions are essential to closing the grid gap for AI factories,” said Sikander Rashid, Global Head of AI Infrastructure at Brookfield. For both groups, this model could define a new standard in a rapidly evolving sector that requires long-term capital commitments.
