Bloom Energy Is Riding the Data Center Boom to a $25 Billion Backlog. Plug Power Is Passing on It. Here's Why.

Bloom Energy secured expanded Brookfield financing totaling $25 billion for AI-related data-center power infrastructure, positioning its on-site solid oxide fuel cells to address rapidly growing electricity demand. Bloom says every major U.S. hyperscaler and more than a dozen neocloud, AI lab, and colocation operators have validated its solutions; analysts rate the shares Moderate Buy with as much as 35% upside to a $354 target. Plug Power, rated Hold, is prioritizing material handling, electrolyzers, and hydrogen production rather than materially pursuing data-center power opportunities, aside from a Microsoft backup-power fuel-cell trial.

Analysis

The key investable distinction is not fuel-cell technology but financing-led speed to deployment. BE can convert data-center power scarcity into contracted project volume without carrying the full balance-sheet burden, potentially improving its sales cycle and lowering customer adoption friction; BAM gains incremental fee-bearing deployment opportunities with infrastructure-like duration. The valuation risk is that a financing framework is being priced as committed equipment backlog: the relevant verification points are executed capacity reservations, project-level returns, gas pass-through terms, and BE's working-capital conversion.

Over the next 1-3 months, BE should retain AI-power narrative momentum if hyperscaler awards or named colocation contracts emerge, while PLUG remains exposed to a less-forgiving profitability test in hydrogen and material handling. PLUG's restraint is not necessarily bearish long term: avoiding capital-intensive, unproven data-center commitments could preserve liquidity, but it leaves the stock without a near-term catalyst capable of offsetting concerns around cash burn and gross-margin execution. A broader second-order beneficiary is distributed-power equipment and gas infrastructure, whereas grid-dependent generation developers risk losing premium data-center load to behind-the-meter solutions.

Contrarian view: BE's addressable market is real, but on-site gas generation can face permitting, emissions, fuel-price and customer-decarbonization constraints that make near-term deployment materially slower than investor expectations. The thesis fails if announced projects do not translate into recognized revenue and improving gross margin over the next two earnings cycles, or if customers favor utility-scale generation plus storage once interconnection timelines improve. BAM is the cleaner, lower-beta expression if the opportunity is primarily an infrastructure-financing bottleneck rather than a sustained equipment-margin inflection.

Key Decisions for Investors

• Initiate a 3-6 month long BE / short PLUG pair, sized beta-neutral: BE has the clearer catalyst path from signed data-center deployments, while PLUG must demonstrate liquidity discipline and hydrogen-business margin recovery. Reassess if BE fails to disclose contracted capacity or if PLUG materially raises liquidity/run-rate guidance.
• Prefer BAM over a standalone BE long for conservative AI-power exposure over 6-18 months. Add on evidence that committed capital is being drawn into operating projects; downside protection comes from BAM's diversified fee base, while the primary risk is that project financing remains non-deployed capacity.
• Do not chase BE solely on financing headlines; use a pullback or wait for the next earnings release to confirm backlog conversion, gross-margin trajectory, and customer prepayment/working-capital terms. A revenue guide increase without corresponding margin or cash-flow improvement would be a reduce signal.
• Maintain PLUG as an avoid/watch rather than an AI-power short catalyst trade over the next quarter. A material data-center order, strategic financing, or explicit reduction in cash burn would invalidate the relative-underperformance thesis quickly.