AI is Driving Utilities to Spend a Record $240 Billion in 2026. Buy These Stocks to Capitalize on the Power Surge.

Utilities may need to spend up to $240B in 2026 to meet AI-driven electricity demand, but regulated rate pass-through is facing pushback as power prices rise. The article highlights Bloom Energy with a $6B fuel-cell backlog (+2.5x YoY) and $20B total backlog, but notes the stock is up >1,000% in a year and trades at ~29x price-to-sales amid no sustainable profits. It suggests comparatively more attractive options—Brookfield Renewable (P/S ~1.6x, ~4.5% distribution yield targeting 5–9% annual growth) and NextEra (22.5x P/E below its 5-year average; proposed Dominion acquisition in Virginia with ~9% earnings growth and ~6% dividend growth).

Analysis

The tradeable distinction here is not “AI power” versus not; it is who can monetize constrained electrons without waiting on rate-case politics. Assets with contracted output or behind-the-meter economics should capture the first wave of incremental spend, while regulated utilities risk a lag where capex rises immediately but earnings recovery trails by quarters. That gap is where equity underperformance can show up even if headline demand stays strong.

The cleanest second-order winner is the cheapest contracted-cash-flow vehicle, not the highest-beta story stock. Brookfield Renewable has the right mix of long-duration contracts and balance-sheet flexibility to absorb demand from hyperscalers, while NextEra offers a lower-volatility way to own the same theme with better diversification across development and regulated assets. Dominion is more of an option on data-center geography than a core thesis, but merger timing and regulatory scrutiny can keep that upside capped in the near term.

Bloom Energy is the obvious “AI power” expression, which is exactly why the risk/reward looks stretched. If backlog converts slower than expected, or if customers defer on-site generation in favor of grid-backed PPAs, the multiple can compress hard because the valuation already discounts a long runway of success. Contrarian view: the market may be overestimating how much value accrues to niche power hardware and underestimating how much accrues to long-duration contracted generators and infrastructure owners.

Catalysts to watch over 1-3 months are state utility rate decisions, hyperscaler PPA announcements, and any evidence that AI capex is being rephased. Over 6-18 months, the key falsifier is whether utility earnings actually keep pace with capex and whether contracted power providers convert bookings into free cash flow rather than just headline backlog.

Key Decisions for Investors

• Long BEP vs. short BE: favor the lower-valuation contracted-cash-flow asset over the high-multiple backlog story; target 3-6 months, with the pair breaking if BE starts converting backlog into durable free cash flow faster than expected.
• Buy NEE on pullbacks as a lower-volatility AI power proxy; 6-12 month horizon, with upside if contracted renewables and data-center demand keep supporting estimate revisions.
• Treat regulated utilities as a hedgeable underweight until rate recovery is clearer; if you need sector exposure, prefer a barbell of BEP/NEE over plain regulated-beta names, and reassess if state-level approvals improve over the next 1-2 quarters.
• Watch D rather than chase it: only become constructive if merger/regulatory path de-risks and Virginia data-center exposure translates into visible earnings accretion; otherwise it remains an option, not a core long.