Bloom Energy Has Found an Unexpected Customer: AI

Bloom Energy (NYSE:BE) designed its solid-oxide fuel cells for corporate campuses, hospitals and retailers. AI hyperscalers have since become the core of its business. The 24/7 Wall St. price target for Bloom is $195.42 over the next 12 months, or 10.95% above the current $176.13. The model rates the stock a buy with high confidence.

Our target sits well below the Wall Street consensus of $284.88. The gap reflects volatility: Bloom has a beta of 3.618, which means sharp swings in both directions, so the model discounts heavily for volatility. Even so, it still points higher from here.

A 40% Slide Followed a Record Quarter

Bloom shares have dropped 20.76% in a week and 40.45% in a month. They remain up 102.72% year to date. The stock trades about 49.9% below its 52-week high of $351.28 and roughly 132.7% up from its low of $75.70. The selloff came after a record earnings report, as investors took profits on a stock where beats were already priced in.

Third-quarter revenue reached $1.24 billion against a $1.07 billion consensus, a gain of 138.1% from a year earlier. Non-GAAP EPS of $0.94 exceeded the $0.67 estimate.

Bloom also raised its full-year outlook for the third time, to $4.4 billion to $4.6 billion in revenue and $3.10 to $3.40 in non-GAAP EPS. Chief executive K.R. Sridhar summed up the demand: “Hyperscalers have been the first to embrace on-site power. They need power rapidly and at large scale.”

Gigawatt Orders Could Drive Shares to $294

In the bull case, Bloom reaches $294.21. Speed to power is the selling point. Bloom says it “delivered power to [Oracle’s] Utah site in less than two months”, and it has servers ready for a 300 megawatt Nebius Group project. American Electric Power disclosed plans for 700 megawatts of carbon-capture-enabled Bloom servers to supply Google.

Bloom’s factories can already produce over five gigawatts a year. Demand has room to keep growing, since Lawrence Berkeley National Laboratory projects data centers could use between 6.7% and 12% of U.S. electricity by 2028.

Bloom is one of a handful of power, cooling, and networking names riding that AI expansion, and we covered seven of them in a free report here. Analyst support is solid, with 5 analysts giving a Strong Buy rating, alongside 10 Buy ratings.

Cash Burn and Valuation Could Pull Shares to $157

In the bear case, the stock falls to $157.23. Bloom trades at 102x trailing earnings, so any slowdown in AI spending would hit the shares hard. Free cash flow was a use of $181 million last quarter. Fourth-quarter results also hinge on “the timing of a small number of deliveries around year-end”.

The cash burn reflects spending on new capacity at Fremont and Newark. Year-to-date free cash flow remains positive $42 million, and Bloom has $2.5 billion in cash. Changes to federal tax credits and slow air permits are still open risks.

Bloom Grows Faster Than Vertiv and Plug Power

Vertiv (NYSE:VRT | VRT Price Prediction) is a useful comparison because both companies sell power gear to AI data centers. Vertiv trades at 24x forward earnings, with quarterly revenue growth of 61.2%. Bloom earns its higher multiple by growing more than twice as fast.

Plug Power (NASDAQ:PLUG) is a direct fuel cell rival, but it posted trailing gross profit of -$184.1 million and revenue growth of just 2.5%. Against these two, our target looks reasonable.

Our Model Rates Bloom a Buy After the Selloff

The 24/7 Wall St. price target of $195.42 comes with a buy rating and 90% confidence. The balance tips because guidance keeps rising while the share price falls.

Keep an eye on whether fourth-quarter deliveries land on time and free cash flow becomes positive again. Cuts to hyperscaler capex would weaken the case. After a 40.45% drop in a month, the valuation risk looks smaller than it did in October.

If current growth and market conditions has, here is where the 24/7 Wall St. model projects Bloom could trade.

Bloom needs to keep executing on its current strategy for these projections to has. Carbon capture adoption or a pullback in AI capex would move results well above or below these levels.

Contact [email protected] for any questions or corrections.