Bloom Energy Stock Rebounds 14% From Its $251 Low After Oracle Confirms the Project Jupiter Contract — $300 Test

Bloom Energy (NYSE: BE) traded at $285.54 late Friday morning, September 25, up $18.89 or 7.08% from Thursday's close of $266.65. The stock opened near the bottom of its range, traded as low as $264.01 in the first minutes of the session, and climbed as high as $286.93 before noon, a swing of $22.92 or 8.7% from low to high. At $285.54, shares sat 0.5% below the session peak and 8.2% above the intraday low. An earlier snapshot at 11:03 a.m. ET showed the stock at $284.30, up $17.65 or 6.62%, on 7.14 million shares.

The move ranks Bloom among the strongest large-cap performers on a day when the S&P 500 gained 0.22%. At $285.54, the company carries a market capitalization of $84.1 billion, based on 294.5 million shares outstanding. The stock has now risen 230% year to date, from a December 31, 2025 close of $86.62, and 279% over the past twelve months. Its 52-week range spans $66.62 to $351.28, placing Friday's price 329% above the low and 18.7% below the high.

Friday's rebound reverses a sharp Thursday selloff. On September 24, Bloom fell as much as 8.7% intraday to $251.30 after reports that Oracle had issued a force majeure notice on Project Jupiter, the New Mexico AI data center campus that Bloom's fuel cells are contracted to power. Volume hit 18.0 million shares, 39% above the stock's average. The shares recovered into the close, finishing down $8.54 or 3.10% at $266.65. From Thursday's intraday low to Friday's high, Bloom rallied $35.63, or 14.2%, in under 24 hours.

The recovery came after Bloom said Oracle remains committed to its contract for 2.4 gigawatts of fuel cell capacity and to the project's planned timeline. That statement did the heavy lifting. Investors who sold on the force majeure headline bought back on the reaffirmation.

The thesis for this forecast runs through every section that follows. Bloom Energy has become the market's purest play on the AI power bottleneck, with record revenue, expanding margins, and a customer list that now includes the largest data center builders in the world. The Oracle scare tested whether that story can survive a delay at its single largest deployment, and the market answered within a day. But at $84.1 billion and more than 100 times this year's projected earnings, the stock is priced for flawless conversion of a multi-gigawatt backlog. Friday's bounce confirms demand for the story. It does not remove the execution risk the Oracle notice exposed.

Project Jupiter: The Force Majeure Notice, the 17-Mile Pipeline, and Bloom's 2.4 GW Reaffirmation

Thursday's selloff traced back to a single document. Oracle sent a force majeure notice to the developer of Project Jupiter, a unit of Blue Owl Capital operating as Stack Infrastructure, related to delays at its Doña Ana County, New Mexico data center. The notice is designed to let Oracle defer certain payments if the campus misses its 2028 operational target. Oracle said publicly that Project Jupiter remains on its planned schedule and that it is fully committed to New Mexico. Blue Owl said the notice does not change the parties' financial commitments. Oracle stock fell 3.47% on Thursday.

The link to Bloom is direct. Oracle partnered with Bloom to power the 2.45-gigawatt campus with solid oxide fuel cells, which generate electricity electrochemically and use minimal water, a critical advantage in New Mexico's arid climate. The campus is intended to supply AI computing capacity for OpenAI. The broader Oracle partnership allows deployment of up to 2.8 gigawatts of Bloom fuel cells, and Oracle initially contracted 1.2 gigawatts of capacity that Bloom expected to deploy within a year.

The vulnerability is the fuel. Bloom's cells require a steady supply of natural gas, and Project Jupiter depends on a 17-mile natural gas pipeline known as the Green Chile Project, operated by Energy Transfer. That pipeline has slipped from its original August 2026 in-service target to February 2027, a six-month delay, following repeated permitting denials. Without the pipeline, Bloom's installed cells cannot run, which leaves the campus without a viable power source and pushes Bloom's revenue recognition on the project to the right. Project debt tied to the development has traded below 90 cents on the dollar, a sign that lenders are pricing a meaningful delay.

Bloom's response on Friday reframed the risk. The company said Oracle remains committed to its 2.4-gigawatt contract and to the planned timeline. Bloom has also argued that its equipment is fungible: fuel cells built for a delayed site can be redeployed to other customers facing grid connection delays. With demand for on-site power outrunning supply across the data center industry, that flexibility matters.

The scale of the exposure explains why the stock moved so hard. Bloom disclosed on its second-quarter call that it has visibility for 25 gigawatts of deployments across its planning horizon. Project Jupiter's 2.45 gigawatts represents 10% of that figure, but a far larger share of the near-term ramp. A single project representing that much of the next two years' deliveries is anchor exposure. Oracle's reaffirmation removes the worst-case reading of the notice. It does not remove the pipeline delay, and investors will be watching the February 2027 in-service date as closely as they watch Bloom's own guidance.

Q2 2026: $1.065 Billion Revenue, 166% Growth, and the First $1 Billion Quarter

Bloom's fundamentals explain why the market bought the Friday dip so aggressively. The company's second-quarter 2026 results, filed on July 28, showed a business in the steepest part of its growth curve.

Revenue reached $1,065.4 million, up 165.5% from $401.2 million a year earlier, marking the first quarter in Bloom's history above $1 billion. Revenue grew 42% sequentially from the first quarter, which implies first-quarter revenue of $750 million. Product revenue, the sale of Energy Servers, rose 215.4% to $935.4 million from $296.6 million and accounted for 88% of total revenue. The shift toward product sales matters because product carries higher margins than the company's service contracts.

Profitability expanded faster than revenue. GAAP gross margin climbed to 33.4%, up 668 basis points from 26.7% a year earlier. Non-GAAP gross margin rose 604 basis points to 34.3% from 28.2%. Gross profit totaled $355.6 million. Operating income swung to $182.2 million from an operating loss of $3.5 million in the second quarter of 2025, a $185.7 million improvement. The GAAP operating margin reached 17.1%. Non-GAAP operating income rose to $239.6 million from $28.6 million, for a 22.5% non-GAAP operating margin.

Earnings per share followed. GAAP EPS came in at $0.62, compared with a loss of $0.18 a year earlier, a $0.80 swing. Non-GAAP EPS reached $0.78, up from $0.10. Cash flow from operations totaled $226.4 million, a $439.5 million improvement from $213.1 million of cash burned in the same quarter of 2025. Cash and equivalents stood at $2.67 billion at June 30.

The margin progression tells the long-term story. Non-GAAP gross margins have expanded from 22% in 2021 to an expected 34% in 2026, reflecting scale benefits and manufacturing efficiencies. Management has emphasized that research and development and general administrative costs are largely fixed, so each additional gigawatt of deliveries carries little incremental overhead. That operating leverage is what turned a $3.5 million loss into $182.2 million of profit in twelve months.

The stock's reaction to the report shows how quickly sentiment has shifted. Bloom closed at $166.84 on July 28, the day of the release, after falling 11.34% during the regular session. It jumped 11.24% after hours to $185.60. From that July 28 close to Friday's $285.54, the stock has gained 71.1% in two months.

2026 Guidance Raised to $3.9–$4.2 Billion: What the Second Half Has to Deliver

The guidance raise on July 28 set the bar that the stock now trades against. Bloom lifted its 2026 revenue outlook to $3.9 billion to $4.2 billion from the prior range of $3.4 billion to $3.8 billion. The midpoint moved from $3.6 billion to $4.05 billion, an increase of $450 million, and represents 100% growth over 2025 revenue of $2.0 billion.

The profit outlook rose alongside it. Bloom now expects non-GAAP operating income of $800 million to $900 million, up from its earlier range, which implies a midpoint operating margin of 21%, compared with a 14% midpoint margin in the company's initial 2026 outlook. Non-GAAP operating income for 2026 would be 3.8 times the 2025 level. Management kept its full-year non-GAAP gross margin outlook at 34%. Non-GAAP diluted EPS guidance stands at $2.55 to $2.85. The company removed its informal free-cash-flow metric from its supplemental presentation.

The math on the second half is demanding. With first-quarter revenue of $750 million and second-quarter revenue of $1,065.4 million, Bloom generated $1.82 billion in the first six months. Reaching the $4.05 billion midpoint requires $2.23 billion in the second half, an average of $1.12 billion per quarter. The low end of $3.9 billion requires $2.08 billion, or $1.04 billion per quarter. The high end of $4.2 billion requires $2.38 billion, or $1.19 billion per quarter. Consensus estimates for the third quarter call for revenue of $1.07 billion and earnings of $0.72 per share, which would sit just above the second-quarter level.

Management built the outlook in two layers. The foundation is backlog conversion: signed commitments delivering against customer site-readiness dates. The second layer is in-year bookings: Bloom intentionally reserves manufacturing capacity for time-to-power customers who need electricity within months, and it expects to book and convert that capacity at rates consistent with recent experience.

That structure is where the Oracle notice bites. If Project Jupiter's site readiness slips with the Green Chile pipeline, deliveries scheduled for that campus shift out of the backlog-conversion layer. Bloom would need to backfill that capacity with in-year bookings from other customers to hold the guidance. Management's argument is that demand from time-to-power customers exceeds available capacity, so redeployment is straightforward. The third-quarter report, due on or around October 27, will be the first test of that claim.

Backlog, Financing, and 25 GW of Visibility: The Machinery Behind the Ramp

Bloom's growth depends on three things beyond its factory: a backlog of signed orders, a pipeline of future deployments, and third-party capital to finance installations. All three expanded sharply in 2026.

In February, Bloom disclosed a total backlog of $20 billion, including $6 billion of product orders. The balance consists of long-term service contracts and other commitments that generate recurring revenue over the life of installed systems. By the second-quarter call in July, management said it had visibility for 25 gigawatts of deployments across its planning horizon, a figure that includes signed contracts, framework agreements, and active negotiations.

The customer base has shifted decisively toward AI infrastructure. Hyperscalers, AI labs, and colocation operators are adopting Bloom systems to bypass grid connection delays that can stretch for years. Chief Executive KR Sridhar has described the shift as secular, calling on-site power a business necessity for AI hyperscalers rather than a slogan. On the July call, he argued that a one-gigawatt data center operated by a full-stack AI provider can generate $12 billion to $24 billion in revenue per year, which makes speed-to-power the dominant economic factor for customers and places the cost of Bloom's equipment in a different context.

Financing is the less visible but equally important piece. Most Bloom deployments are funded by third-party capital rather than by the end customer's balance sheet. Brookfield has expanded its project financing framework with Bloom to up to $25 billion. Industrial Development Funding, which has previously funded Bloom deployments, partnered with Oaktree, MUFG Bank, and Morgan Stanley to bring its cumulative commitment to $2.6 billion, with more partners in negotiation. Those arrangements let Bloom sell equipment and recognize product revenue while financiers own the installed assets and collect payments over time.

The financing channel is where higher interest rates matter most. Every deployment funded by a financier carries a cost of capital that tracks Treasury yields and credit spreads. With the 10-year Treasury at 5.2%, the highest since 2007, the hurdle rate for those projects has climbed. So far, the economics of AI data centers have absorbed that cost, because the revenue per gigawatt dwarfs the financing expense. If rates keep rising and project debt trades lower, as it has on Project Jupiter, financiers may demand higher returns or slow commitments. The backlog is large, but its conversion runs through capital markets that are under visible stress.

S&P 500 Entry on September 21: Index Buying, Then a Sell-the-News Test

Bloom joined the S&P 500 before the market opened on Monday, September 21, replacing Molson Coors Beverage. Index inclusion forces passive funds tracking the benchmark to buy the stock, and anticipation of that buying drove much of the September rally. The stock gained 31.86% in the month through September 21 and is on track for its best month since April.

The price path around the inclusion followed a familiar pattern. Bloom closed at $272.89 on September 21, up 2.73% on the first day as an index member, outperforming the S&P 500's 1.49% gain. On September 22, it rose 1.49% to $276.97, touching an intraday high of $284.25 on 10.56 million shares, while the Nasdaq Composite set a record close of 27,244.28. On September 23, as the 10-year Treasury yield jumped to 5.135% and the Nasdaq fell 1.13%, Bloom slipped to $275.19. Then came the Oracle notice on Thursday.

The timing of Thursday's drop was not an accident. Market participants often sell a stock once an anticipated event materializes, and index inclusion is one of the most anticipated events in equity markets. Hedge funds that bought ahead of the September 21 effective date had a built-in exit once passive funds completed their purchases. The Oracle headline gave them a reason to sell. The stock's 8.7% intraday drop on Thursday combined sell-the-news selling with genuine concern about Project Jupiter.

Volume confirms the shift in participation. The stock's three-month average daily volume stands at 16.29 million shares, while a separate average measure puts it at 12.92 million. Thursday's 18.0 million shares traded 39% above the lower average. Friday's 7.14 million shares by 11:03 a.m. put the stock on pace for a full-day total near its average, a sign that the rebound came on steady buying rather than a panic squeeze.

Index membership changes the stock's shareholder base permanently. Passive funds now hold a fixed percentage of Bloom's float that will not trade on news, which reduces the free float available to active investors and can amplify moves in both directions. It also puts Bloom in front of every large-cap portfolio manager benchmarked to the S&P 500, a pool of capital that had little reason to own a $20 billion company two years ago but must now justify being underweight an $84 billion one.

The Scandium Class Action: A February 2025–July 2026 Class Period and a September 28 Deadline

A second legal overhang sits alongside the Oracle notice. A securities class action captioned Nevins v. Bloom Energy Corporation, No. 26-cv-07944, is pending in the U.S. District Court for the Northern District of California. It charges Bloom and certain top executives with violations of the Securities Exchange Act of 1934 on behalf of investors who bought shares between February 27, 2025 and July 8, 2026. The deadline for investors to seek appointment as lead plaintiff is Monday, September 28, 2026.

The case centers on scandium, a rare-earth metal used in Bloom's solid oxide fuel cells. The complaint alleges that Bloom obtained scandium through intermediaries who sourced the metal from China and, as a result, understated the extent of its reliance on Chinese scandium. It further alleges that Bloom's positive statements about its business, operations, and prospects were materially misleading as a consequence. A related claim contends that Bloom understated its exposure to Chinese export controls and U.S. tariffs by misrepresenting where critical raw materials and components originated.

The trigger was a July 8, 2026 report, published at 1:00 p.m. EST, alleging that Bloom relied on Chinese scandium based on global trade data, Chinese corporate filings, satellite imagery, and messages with suppliers. Bloom shares fell $15.28, or 5.7%, to close at $254.29 that day on heavy volume. Friday's price of $285.54 sits 12.3% above that July 8 close, which means the stock has more than recovered the damage.

The legal risk is modest in dollar terms and significant in operational terms. Securities class actions of this kind typically take years to resolve and often settle for amounts that are small relative to a company's market value. The larger question is supply-chain exposure. China has used export controls on rare-earth metals as a trade weapon, and scandium supply is concentrated. If Bloom's supply chain depends on Chinese scandium, any tightening of Chinese export policy could constrain its ability to ramp production toward its 2026 guidance. Trade tensions remain live: this week's Trump-Xi state visit produced an agreement to keep the trade relationship at the status quo for several months, but no structural resolution.

Monday's lead-plaintiff deadline is a procedural event and unlikely to move the stock on its own. The lead plaintiff, once appointed, will file an amended complaint that could add detail on the sourcing allegations. That filing is the next point at which the issue could resurface.

Valuation: 106 Times 2026 Earnings, 21 Times Sales, and a P/E Above 300

Bloom's valuation is the core tension in any forecast for the stock. At $285.54 and 294.5 million shares, the company's market capitalization is $84.1 billion.

On trailing earnings, the stock trades at a price-to-earnings ratio of 310, a figure that reflects the loss-making quarters still in the trailing twelve months. Earlier in September, with the stock near $266, trailing P/E measures stood between 357 and 365. Trailing multiples overstate the valuation of a company whose profitability has only just turned, so forward measures carry more weight.

On 2026 guidance, the picture is still demanding. The non-GAAP EPS range of $2.55 to $2.85 has a midpoint of $2.70. At $285.54, the stock trades at 105.8 times midpoint 2026 earnings, 112 times the low end, and 100.2 times the high end. On revenue, the $4.05 billion midpoint implies a price-to-sales ratio of 20.8. On operating profit, the $850 million midpoint of non-GAAP operating income implies a market-cap-to-operating-income multiple of 99.

For context, those multiples sit at the extreme end of the S&P 500. A forward P/E above 100 is typically reserved for companies with multi-year growth runways and near-certain execution. Bloom has the runway, with 25 gigawatts of visibility and revenue doubling in 2026. The execution certainty is what the Oracle notice called into question.

The bull argument on valuation rests on the second derivative. If Bloom doubles revenue again in 2027 to $8 billion and expands non-GAAP operating margins from 21% toward 25%, operating income would reach $2 billion. At that level, today's $84.1 billion market cap would represent 42 times 2027 operating income, a multiple that looks less extreme for a company growing 100% per year. The consensus price target of $330, raised from $290 earlier this month, implies 15.6% upside from Friday's price.

The bear argument rests on the discount rate. With the 10-year Treasury at 5.18%, the present value of earnings expected in 2028 and beyond falls with every basis point of yield. High-multiple growth stocks are the most sensitive to that math. A company priced at 106 times current-year earnings needs its long-dated cash flows to arrive on schedule, and Project Jupiter is exactly the kind of long-dated deployment whose timing just became uncertain.

5.2% Treasury Yields and Project Finance: The Macro Headwind for Capital-Heavy Growth

Bloom's business model sits at the intersection of two forces pulling in opposite directions this week: accelerating AI capital spending and the highest long-term interest rates in two decades.

The capital spending side is strong. The Census Bureau's durable goods report on Friday showed orders for nondefense capital goods excluding aircraft, the core measure of business investment, jumping 1.6% in August against a 0.5% forecast. Electrical equipment orders rose 1.1%, and computer orders gained 1.5%. Microsoft projected capital expenditures above $50 billion for a single quarter and $175 billion for calendar 2026. Akamai signed an $11.6 billion, seven-year cloud infrastructure deal with Anthropic on Thursday that requires $5.5 billion of related capital spending. Every one of those builds needs power, and grid connections in major U.S. markets can take years.

The rate side is hostile. The 10-year Treasury yield hit 5.225% on Thursday, its highest level since July 2007, and traded between 5.169% and 5.209% on Friday. The 30-year bond yield touched 5.502%, the highest since 2004. The Federal Reserve raised its benchmark rate to 3.75% to 4.00% on September 16, and futures price a 66% to 71% probability of another hike in October. Those rates flow directly into the cost of project financing for Bloom installations, most of which are owned by third-party financiers who borrow to buy the equipment.

The Oracle situation shows where those forces collide. Project Jupiter's debt trading below 90 cents on the dollar reflects both the pipeline delay and the higher cost of capital: a delayed project in a high-rate environment accrues more interest before it generates cash. If more projects slip, financiers will demand higher returns, which raises the effective cost of Bloom's systems to end customers.

Bloom's week-by-week trading has tracked yields. On September 23, when the 10-year jumped to 5.135%, the stock slipped while the Nasdaq fell 1.13%. On Friday, when yields stabilized and oil eased on reports of a potential U.S.-Iran deal to reopen the Strait of Hormuz, the stock rallied 7%. Natural gas prices also matter: Bloom's cells run on gas, and Henry Hub futures fell 5.25% to $3.124 on Friday, which lowers the operating cost of every installed system.

The net effect is a stock that trades on AI demand in the medium term and on the 10-year yield in the short term. A move in the 10-year back below 5% would ease the financing pressure on Bloom's backlog. A push toward 6% would test whether AI customers can absorb the higher cost of speed.

Sector and Competition: Fuel Cells Versus Turbines, Engines, and the Grid

Bloom competes for the same customers as every other on-site power provider, but its technology gives it a distinct position in the data center market.

The main alternatives are natural gas turbines, reciprocating gas engines, and grid connections. Gas turbine manufacturers such as GE Vernova have order books stretching years into the future as utilities and data center developers race to add generation, which limits how quickly new turbines can be delivered. Reciprocating engine makers, including Caterpillar and Cummins, offer faster deployment but lower efficiency and higher emissions. Grid connections remain the cheapest option per megawatt-hour when available, but interconnection queues in major data center markets can run for several years. Bloom's pitch is speed: its modular fuel cell systems can be installed in months, scaled in increments, and operated with minimal water.

The water advantage is concrete. Solid oxide fuel cells generate electricity electrochemically rather than through combustion, so they avoid the cooling-water demands of steam-cycle generation. That made Bloom the choice for Project Jupiter in arid New Mexico, and it gives the company an edge in water-constrained regions across the southwestern United States, the Middle East, and parts of Asia.

The natural gas dependency is the flip side. Bloom's cells need a steady gas supply, and Project Jupiter's pipeline delay shows that fuel infrastructure can become the binding constraint even when the power equipment is ready. Turbines and engines face the same dependency, so this is not a competitive disadvantage against other gas-fired options, but it does limit Bloom's claim to fully bypass infrastructure bottlenecks.

Among fuel cell peers, Bloom has pulled far ahead. Plug Power traded at $1.99 on Friday, up 1.65% on 22.6 million shares, with a market capitalization of $2.78 billion and a 52-week decline of 17%. Bloom's $84.1 billion market cap is 30 times Plug's. FuelCell Energy remains a small-cap player. The divergence reflects a market that has picked a winner in stationary fuel cells for data centers: Bloom's solid oxide technology and manufacturing scale have made it the default choice for hyperscalers.

The broader AI power trade moved with Bloom on Friday. Contract manufacturers Jabil and Flex gained 4.37% and 4.26%, Dell rose 4.76%, and Akamai jumped on its Anthropic deal. Bloom's 7% gain made it one of the strongest names in a group that the market treats as a single thematic bet on AI infrastructure spending.

Resistance Map: $286.93, $300, the $330 Target, and the $351.28 52-Week High

The overhead levels for Bloom are spaced widely, reflecting the stock's volatility and the distance from its prior peak.

The first resistance is $286.93, Friday's intraday high. The stock traded within $1.39 of that level at $285.54, and a close above it would mark the highest finish since the stock's earlier run toward its 52-week peak. Sept 22's intraday high of $284.25 sits just below it, forming a two-day resistance band between $284 and $287. That band capped the stock on the first full day of S&P 500 membership and again on Friday morning. A daily close above $287 would confirm the breakout from the post-inclusion range.

The second level is $300, a psychological round number 5.1% above Friday's price. Crossing $300 would lift the market capitalization to $88.4 billion. The stock traded above $300 earlier in the year on its run to the 52-week high, and prior buyers at that level represent potential supply.

The third level is the $330 consensus price target, raised from $290 earlier in September. It sits 15.6% above Friday's price and would put the market cap at $97.2 billion. Reaching it before the October 27 earnings report would require a combination of falling yields, a clean resolution of the Project Jupiter timeline, and continued AI infrastructure momentum.

The fourth and final resistance is the 52-week high of $351.28, 23.0% above Friday's price. At that level, Bloom's market cap would reach $103.5 billion, and its forward P/E on midpoint 2026 guidance would climb to 130. A return to the high would likely require evidence that 2027 revenue can double again, most likely from a third-quarter report that exceeds consensus and a raise in full-year guidance.

The technical structure supports a test of the first two levels. Bloom's recovery from its July 28 close of $166.84 has formed a sequence of higher lows, and the stock's recent consolidation above key moving averages has taken the shape of a flag, a continuation pattern in an uptrend. A flag breakout typically targets a move equal to the preceding advance, which in Bloom's case would point well above $300. That target depends on the pattern completing, and Thursday's drop to $251.30 came close to breaking it.

Support Map: $275, $266.65, $254.29, $251.30, and the $166.84 July Base

The downside levels for Bloom are closer than the upside ones and carry specific event histories.

The first support is the $275 area, marked by the September 23 close of $275.19 and the September 21 to 22 range between $272.89 and $276.97. That zone held for three sessions before the Oracle notice broke it. A retest from above on any pullback would confirm it as support. From $285.54, $275 is 3.7% lower.

The second support is $266.65, Thursday's close and Friday's reference price. The stock opened Friday just below that level at $264.01 and reclaimed it within minutes. A daily close below $266 would erase Friday's rebound entirely and signal that the Oracle reaffirmation failed to hold buyers. It sits 6.6% below the current price.

The third support is $254.29, the July 8 close after the scandium sourcing report. That level marks the bottom of the event that triggered the class action and represents an anchor for investors who bought the July drop. It is 10.9% below Friday's price.

The fourth and most important short-term support is $251.30, Thursday's intraday low on the Oracle headline. It is 12.0% below the current price. That low defines the market's first reaction to the Project Jupiter risk. A break below it would indicate that investors see the delay as a genuine threat to 2026 guidance rather than a timing shift, and it would likely trigger stop-loss selling from traders who bought the Thursday dip.

Below $251.30, the stock's post-earnings base sits far lower. The July 28 close of $166.84 marked the pre-earnings low, and the after-hours rebound to $185.60 marks the gap that opened on the second-quarter report. A return to that zone would represent a 35% to 42% decline and would require a combination of a guidance cut, a broad AI spending slowdown, and a 10-year Treasury yield approaching 6%.

The support map shows how much Friday's rebound matters. Bloom is 12.0% above Thursday's panic low and 7.1% above Thursday's close, a cushion that a single negative headline on Project Jupiter or the scandium case could erase.

Bloom Energy Forecast: $266–$300 Base Range Into October 27 Earnings, $330 Upside, $251 Downside Risk

The forecast for Bloom Energy over the next five weeks turns on three dated events: the September 28 class action lead-plaintiff deadline, the Federal Reserve's late-October meeting, and the company's third-quarter earnings report on or around October 27.

The base case, carrying the highest probability, is a range between $266 and $300. Thursday's close of $266.65 defines the floor, and the $300 round number defines the ceiling. In this scenario, Oracle's reaffirmation of the 2.4-gigawatt contract holds, the Green Chile pipeline stays on its February 2027 schedule, the 10-year Treasury yield holds between 5.0% and 5.3%, and the stock consolidates its September gains ahead of earnings. Friday's price of $285.54 sits in the middle of that range.

The bullish scenario targets $330, a gain of 15.6%. It requires a third-quarter revenue print above the $1.07 billion consensus, confirmation that deliveries tied to Project Jupiter can be redeployed to other customers without affecting 2026 guidance, and a raise in the full-year revenue range above $4.2 billion. A drop in the 10-year yield below 5%, most likely on a confirmed U.S.-Iran deal that lowers oil and inflation expectations, would add to it by easing the financing cost of Bloom's backlog. A close above $287 would confirm the breakout, and a close above $300 would open the path to the target. Beyond $330, the 52-week high of $351.28 becomes the next objective.

The bearish scenario targets $251, a decline of 12.0%. The trigger would be a further delay in the Green Chile pipeline beyond February 2027, a disclosure that Project Jupiter deliveries are being deferred out of 2026, an amended class action complaint that adds detail on Chinese scandium dependency, or a 10-year yield breaking above 5.3%. A daily close below $266 would put Thursday's $251.30 low in play. A break of that level would expose the $254 to $251 zone that marks both the scandium-report close and the Oracle-notice low.

The verdict is constructive with elevated execution risk. Bloom has delivered the strongest financial turnaround in the clean energy sector: revenue more than doubling, gross margins up 668 basis points, operating income swinging from a loss to $182.2 million, and guidance that implies 100% growth this year. The market's 14.2% rebound from Thursday's low shows that investors still want exposure to the AI power bottleneck and view Bloom as the leading way to own it. The valuation leaves no cushion for delay. At 106 times midpoint 2026 earnings and $84.1 billion in market value, Bloom needs its backlog to convert on schedule, its supply chain to stay clear of Chinese export controls, and its financing partners to keep funding at 5% Treasury yields. The October 27 report will show whether the Project Jupiter slip is a scheduling footnote or the first crack in the 25-gigawatt story.

That's TradingNEWS