Can Bloom Energy build them?

3 days ago

Bloom Energy's backlog surged to around $20 billion, up 65% year-over-year, with major deals from Oracle, AEP, and Brookfield, despite having shipped only 1.5 GW in its 23-year history.
Demand is driven by hyperscalers willing to pay a premium to avoid lengthy grid interconnection queues, with fuel cells offering deployment in 12-18 months compared to years for alternatives.
Scaling manufacturing from 1 GW/year to over 4 GW/year poses challenges, including ceramic production complexity, a key supplier (MTAR) scaling up, and reliance on scandium from China, which faces export controls.
Fuel cells are capital-intensive ($7,000-$8,000/kW) but competitive when considering faster deployment and levelized costs, especially with federal tax credits, though they depend on natural gas and face gas pipeline bottlenecks.
Bloom leads the stationary SOFC market with ~60% share, facing minimal competition at scale, but must manage risks like customer concentration, execution in ceramic manufacturing, and degradation at gigawatt-scale deployments.
Financials show inflection with first GAAP-profitable quarter in Q1 2026, but high valuation (25x price-to-sales) and reliance on a few customers (e.g., Oracle) pose risks if scaling or contracts falter.
Fuel cells serve as a bridge for AI-driven power demand until grid capacity improves, with Bloom's success hinging on scaling manufacturing quickly to capture this interim market.

Hasty Briefsbeta