For years, Bloom Energy (NYSE:BE) was the kind of clean-energy stock people bought on hope. That’s no longer true. In the second quarter of 2026, the company posted GAAP operating income of $182.2 million, a year after losing $3.5 million in the same quarter.
Bloom builds solid oxide fuel cells, large boxes that turn natural gas or hydrogen into electricity on site without burning it. Customers buy the systems and pay for ongoing service. Right now, its most eager buyers are AI data centers that can’t wait years for a grid connection.
While the turnaround is real, assessing whether it belongs alongside the 12 High-Flying Stocks to Invest In raises a harder question: can a stock priced far above its sector keep earning that premium while the business scales this fast? The answer depends less on demand than on something duller: margins.
Speed Is the Moat
Bloom’s edge is time. A data center that needs power now can’t sit in a utility queue, and Bloom’s systems go in right beside the servers. That pitch is winning approvals. Management says every major US hyperscaler, plus more than a dozen US neoclouds, AI labs, and colocation operators, has approved its systems for AI facilities.
Approval isn’t the same as an order. But it’s the gatekeeping step slower rivals still have to clear, and it places Bloom among the 10 Best AI Energy Stocks to Buy in 2026 as investors evaluate the AI buildout on electricity rather than chips.
Demand Turned Into Cash
The second quarter showed what that demand looks like once it converts. Revenue reached $1,065.4 million, up 165.5% from a year earlier, while product revenue jumped 215.4%. Non-GAAP gross margin (which strips out items like stock-based pay) widened to 34.3% from 28.2%.
Cash finally followed profits. Operating cash flow came in at $226.4 million, compared with $213.1 million burned in the year-ago quarter. Management then raised its full-year 2026 outlook to $3.9 billion to $4.2 billion in revenue and $800 million to $900 million in non-GAAP operating income.
Bloom is also trying to dig the moat deeper. Last month, it published a report arguing its fuel cells can feed 800-volt DC power directly into data centers, the architecture Nvidia has specified for its 2027-and-later rack designs. That ties Bloom to the same shift reshaping Nvidia’s next-generation AI racks. By Bloom’s own model, the approach could cut non-compute capital spending at a 1-gigawatt AI data center by $3.6 billion, or 27%. Those are company estimates, not field results, but they show Bloom pitching itself as part of the data center’s blueprint, not a bolt-on generator.
Can Margins Survive the Growth Spurt?
The real worry is execution. Growing this fast means more factory capacity, more hiring, and more installations at once, and any of those can pinch margins. Full-year guidance calls for a non-GAAP gross margin of about 34%, in line with the second quarter, so there’s little cushion built in.
A stock priced for near-flawless delivery doesn’t need a disaster to stumble; one quarter where costs outrun sales would do it. Bloom’s demand is also concentrated in one booming industry, and if AI spending cools, so does its pipeline. The counterpoint is that the second quarter showed margins widening, not shrinking, as volumes surged.
Pricey, but Cheaper Than Its Past
At 103.62 times forward earnings, as of October 9, Bloom trades far above the sector’s 18.75. That’s a multiple few traditional power equipment makers ever command. If you’re paying over $100 for each dollar of expected profit, the business needs to keep compounding.
History softens the picture. Bloom’s five-year average P/E is 134.54, so today’s multiple sits below its own norm, and this time it’s attached to a company generating real earnings and cash.
Growth is the case for the price tag. Analysts expect earnings per share to climb 81.98% in 2027. If Bloom gets anywhere close, the forward multiple compresses quickly as profits catch up.
One caveat: management guides on adjusted numbers. Its full-year non-GAAP EPS outlook is $2.55 to $2.85, and second-quarter non-GAAP diluted EPS of $0.78 compared with $0.62 on a GAAP basis. This analysis leans on the adjusted figures because that’s how the company frames its targets, but GAAP earnings run lower. Hedge fund interest rose, with 116 funds holding the stock in the most recent quarter, up from 91 in the prior one. Short interest sits at 7.34% of the float, a noticeable but not extreme bearish bet.
From Story Stock to Earnings Stock
Bloom has crossed from story to substance, and the valuation looks steep against its sector but reasonable against its own history. For growth investors who can stomach volatility and heavy exposure to AI spending, the setup is far more defensible than when the company was losing money. What would change that is margins. If gross margin slips meaningfully below the mid-30s while growth slows, the premium would be hard to justify.
READ NEXT: This 6.3% Dividend Yield Looks Too Good to Ignore. Is the Stock a Buy? and 10 Best 52-Week Low Technology Stocks to Buy According to Analysts.
Hedge Fund Resource Center