Bloom Energy makes the box that lets a hyperscaler put a 100 MW data center on a piece of land that does not yet have a grid connection.
Bloom Energy makes the box that lets a hyperscaler put a 100 MW data center on a piece of land that does not yet have a grid connection. The product is the Bloom Energy Server, a solid oxide fuel cell that runs on natural gas, biogas, or hydrogen and converts it directly to electricity through a non-combustion electrochemical reaction. Customers buy the boxes (or have them financed by infrastructure capital), Bloom installs them, then collects long-dated service revenue keeping them running for 15+ years.
The reason BE matters in 2026, after fifteen years of being a niche on-site power company with negative gross margins, is that grid interconnect queues for new US data centers run four to seven years, and AI capex cannot wait that long. Anyone building a gigawatt-scale AI factory now needs behind-the-meter generation that ships in months, not megawatts that arrive in 2031. Bloom is one of three companies (alongside reciprocating gas gensets and small modular reactors that do not yet exist commercially) that can credibly fill that gap. It is the only one operating at gigawatt scale today.
| Segment | What it is | Approx. % |
|---|---|---|
| Product | Energy Server hardware sales (the box itself) | ~74% |
| Service | Long-term O&M contracts (fixed-price, 15+ year terms) | ~14% |
| Installation | Site engineering and commissioning | ~10% |
| Electricity | Managed-services / PPA-style revenue | ~2% |
Source: Q4/FY2024 earnings release. Exact split for FY2025 ($2.02B revenue) not yet broken out at line-item level in the press release.
Business model: Hardware-plus-recurring-service. Each Energy Server sale drags a multi-decade service contract behind it. The economic logic (and historical pain point) is that Bloom records the service revenue but also bears the warranty replacement cost, so the service line was loss-making for years. Margin recovery has been the operational story since 2024.
Geographic mix: US-dominant. Korea is the second meaningful market, served exclusively through SK ecoplant. India, Japan, and Europe are smaller. Detailed split not disclosed at segment level.
Latest investor presentation: Q4 2025 earnings deck (Feb 6, 2026) and the July 2025 Investor Presentation.
Capacity guidance: management says the company is on track to double annual production capacity to 2 GW by end of 2026 (Utility Dive).
SK ecoplant (Korea — primary strategic partner since 2018). SK ecoplant is the environmental and energy-solutions arm of Korea’s SK Group, formerly SK E&C. They are Bloom’s exclusive distributor in South Korea and were a 10%+ shareholder. Cumulative equity investment hit ~$566M, including original $255M into zero-coupon non-voting redeemable convertible preferred at $25.50/share in 2021. The 2023 expanded offtake locked in 500 MW through 2027 worth roughly $1.5B in product revenue plus $3B in 20-year service revenue (Dec 2023 release). In July 2025 SK ecoplant trimmed its position by selling 10M shares via OTC block trade at $27.60 for ~$276M proceeds, but did not exit (Korea Times).
Oracle (April 2026 expansion). Master services agreement for up to 2.8 GW, with the initial 1.2 GW already contracted and being deployed. Oracle also received a warrant on April 9, 2026 to buy up to 3.53M shares at $113.28 (~$400M strike value), expiring October 9, 2026 (CNBC). This is the deal that has reset the BE narrative from “niche fuel cell stock” to “AI infrastructure pick-and-shovel.”
American Electric Power (AEP). Announced November 2024: framework deal for up to 1 GW, ~$2.65B over the term. As of January 2026, the first 100 MW order has been placed and AEP is calling it the world’s largest commercial fuel-cell procurement (Fuel Cells Works).
Brookfield (October 2025). Brookfield committed up to $5B in infrastructure capital to deploy Bloom systems across global AI factory sites (Bloom press release). Stock spiked 20% on the announcement.
Equinix. Decade-long customer that started in 2015 with a 1 MW pilot. February 2025 expansion took deployments past 100 MW across 19 IBX data centers, with ~75 MW operational and ~30 MW under construction.
| # | Customer | Ticker | Est. Revenue Share | Relationship |
|---|---|---|---|---|
| 1 | SK ecoplant | private (SK Group) | Historical 10%+ related-party customer | Exclusive Korea distributor + shareholder |
| 2 | Oracle | NYSE: ORCL | Largest forward backlog driver | Up to 2.8 GW frame, 1.2 GW contracted |
| 3 | American Electric Power | NASDAQ: AEP | Material once 1 GW frame draws | 1 GW frame, 100 MW initial order placed |
| 4 | Equinix | NASDAQ: EQIX | Reference customer, recurring | >100 MW across 19 IBX sites |
| 5 | Brookfield Asset Management / Infrastructure | NYSE: BAM, BIP | Capital partner not a direct revenue line | $5B AI deployment program |
Concentration risk: Historically high. SK ecoplant was a disclosed 10%+ customer in past 10-Ks. With the Oracle frame absorbing capacity through end-2027 per analyst commentary, single-customer concentration is rising not falling. The shape of the risk is changing too: from one strategic partner (SK ecoplant) to one hyperscaler (Oracle) to one infrastructure capital provider (Brookfield) all simultaneously. Top-1 and top-5 % from FY2024 10-K: not pulled in this session.
Dependency flags: Oracle has a $400M warrant struck below the current price, which means Oracle is also a future shareholder. That is unusual: a major customer with a warrant-based equity stake aligns interests but creates a dual relationship that needs disclosure scrutiny.
Bloom solves a problem that did not exist at scale until the AI capex cycle began: how to put 100 MW to 1 GW of clean firm power on a specific piece of land in 12 to 24 months when the local grid cannot deliver it for half a decade. Behind-the-meter SOFC is one of the few technologies that hits the speed and scale requirement today.
End-use applications (FY2026 narrative):
TAM framing. Goldman, JPMorgan and others now cite a US data center power TAM in the $200B–$500B range over the next decade depending on how AI compute demand scales. Bloom’s serviceable slice is the “off-grid or behind-the-meter, fast deploy” subset, which is smaller but where it has near-monopoly positioning at gigawatt scale. Specific company-cited TAM figures in the latest deck: not pulled in this session.
Secular tailwinds: 1. Grid interconnect timelines getting longer, not shorter 2. AI compute power demand outpacing utility-scale generation buildout 3. Hyperscalers willing to pay a premium for time-to-power 4. Natural gas as a transition fuel (US has it cheap and abundant) 5. Eventually, clean hydrogen optionality on the same hardware
The risk to the TAM thesis: if interconnect queues clear, or if SMRs commercialize on schedule, or if hyperscalers redesign for grid-only, Bloom loses its time-to-power moat. None of these are near-term threats but all three are real.
| Name | Title | Tenure | Background |
|---|---|---|---|
| KR Sridhar | Founder, Chairman & CEO | Founder (2001); CEO 25 years | Former director of NASA Ames Center for Space Power; the technology came out of his Mars-mission work on solid oxide cells. Has not stepped into Executive Chair role despite age and tenure. Holds combined Chair/CEO role. |
| Simon Edwards | CFO | April 13, 2026 (new) | Joined from Groq, where he was CFO and then briefly CEO. (Bloom press release) |
| Aman Joshi | Chief Commercial Officer | Multi-year | Owns the customer relationships including the Oracle and AEP wins. |
| Satish Chitoori | Chief Operating Officer | Multi-year | Manufacturing scale-up lead. |
| Maciej Kurzymski | Chief Accounting Officer | Multi-year | Served as interim CFO May 2025 through April 2026. |
| (Departed) Greg Cameron | Former CFO 2020–May 2024 | — | Now CFO at CF Industries. |
| (Departed) Daniel Berenbaum | Former CFO Apr 2024–May 2025 | — | Departed quickly, less than 14 months in role. |
The CFO churn is the loudest governance signal. Three CFOs in roughly two years (Cameron → Berenbaum → Kurzymski as interim → Edwards) is unusual for a company of this size in a complex capital-stack environment. It is not a smoking gun by itself but it warrants scrutiny on financial reporting controls and on whether the underlying issue is strategic disagreement, performance, or something else. Worth probing in the next /mgmt-dd pass.
| Name | Role | Independent? | Background |
|---|---|---|---|
| KR Sridhar | Chair (combined with CEO) | No | Founder |
| Jeffrey Immelt | Lead Independent Director (since May 2020) | Yes | Former Chairman/CEO of GE; brings industrial scale-up credibility |
| Jim Snabe | Director (since August 2025) | Yes | Chairman of Siemens AG; appointment signals strategic positioning toward European industrial customers |
| John T. Chambers | Director | Yes | Former Chairman/CEO of Cisco; brings hyperscaler/networking customer Rolodex |
| Mary K. Bush | Director | Yes | Long-tenured |
| Michael Boskin | Director | Yes | Former Chair of US Council of Economic Advisers |
| Cynthia Warner | Director | Yes |
10-member board total. Specific committee chairs (Audit, Comp, Nom/Gov) not pulled from latest DEF 14A (filed April 2, 2025) in this session.
Direct competitors in on-site / behind-the-meter power for AI data centers:
| Competitor | Technology | Status |
|---|---|---|
| Caterpillar (CAT) / Cummins (CMI) | Reciprocating gas gensets | Off-the-shelf, fast deploy, but burn-cycle emissions and lower efficiency |
| GE Vernova (GEV) | Aeroderivative turbines, hydrogen-ready | Larger scale, longer lead times, more grid-integrated |
| Mitsubishi Heavy / Siemens Energy | Gas turbines | Same |
| NuScale, Oklo, X-energy | Small modular reactors | Not commercial yet; relevant 2030+ |
| PowerSecure / CoreWeave-affiliates | Engineering-procurement-construction firms bundling generation | Emerging |
| Plug Power (PLUG), FuelCell Energy (FCEL) | Other fuel cell types (PEM, MCFC) | Smaller scale, weaker economics, distressed balance sheets |
Bloom’s moat: 1. Operating fleet at gigawatt scale — 1.5+ GW deployed, 15+ years of reliability data. Hyperscaler procurement teams risk-rank vendors on operating history; nobody else in the SOFC space is close. 2. Vertical integration — designs and manufactures the cells, the stacks, the system, and the service model. Most peers outsource one or more layers. 3. Customer reference list — Equinix, Oracle, AEP, SK. That is a reference set that takes a decade to build. 4. Time-to-power — the structural feature that matters most right now. SOFC deploys in months and produces firm power on natural gas immediately.
Competitive vulnerabilities: - Fuel cells run on natural gas in 2026 reality, not hydrogen. Long-term decarbonization story depends on H2 economics that are not yet there. - Reciprocating gensets are cheaper per MW upfront. Bloom’s pitch is lifecycle cost, not nameplate. - Service-segment economics are still the weakest line in the model.
Porter’s Five Forces snapshot: - Buyer power: Moderate-high. Hyperscalers are concentrated and price-sensitive but also speed-sensitive; right now speed is winning. - Supplier power: Moderate. Yttria-stabilized zirconia and other cell materials have multiple suppliers; nickel and ceramics processing is the bottleneck. - New entrants: Low near-term. SOFC manufacturing requires fifteen years of process knowledge. Korean and Chinese state-backed entrants are real long-term threats. - Substitutes: Reciprocating gensets (now), SMRs (later), grid-attached generation (always). - Rivalry: Moderate. The on-site SOFC space is a near-monopoly. The broader on-site power space is more competitive.
Valuation (as of April 27, 2026 close, $234.68)
| Metric | Value |
|---|---|
| Market cap | $66.7B |
| Enterprise value | $66.4B |
| P/E (TTM) | N/A (negative EPS) |
| Forward P/E | 75.2x |
| Price / Book | 85.5x |
| Price / Sales (TTM) | 33.0x |
| EV / Revenue (TTM) | 32.8x |
| EV / EBITDA (TTM) | 477.6x |
| FCF yield | ~0.09% (FCF $57M on $66.7B mkt cap) |
| Dividend yield | 0% |
| 52-week range | $16.05 – $242.20 |
| Beta | 3.19 |
The valuation is the single most important fact about this stock. BE trades at multiples that price in flawless execution against the mid-to-high analyst growth case for 3+ years. Forward EV/Revenue of ~12x on FY2027 consensus revenue ($5.2B) is more reasonable, but you are paying for the consensus to be right.
Income statement & margins
| Metric ($M) | FY2022 | FY2023 | FY2024 | FY2025 | FY2026E | FY2027E |
|---|---|---|---|---|---|---|
| Revenue | 1,201 | 1,335 | 1,474 | 2,024 | 3,227 | 5,237 |
| Revenue growth (YoY) | +23% | +11% | +10.5% | +37.3% | +59% | +62% |
| Gross margin | ~13% | ~15% | ~27.5% | ~29.0% | ~32% guide | n/a |
| Operating income | (261) | (209) | 23 | 73 | 425–475 (guide) | n/a |
| Operating margin | -22% | -16% | +1.6% | +3.6% | ~14% guide | n/a |
| Net income | (301) | (302) | (29) | (88) | n/a | n/a |
| Diluted EPS | ($1.62) | ($1.42) | ($0.13) | ($0.37) | $1.40 | $3.12 |
The story in this table is the operating-income inflection. After years of bleeding, Bloom flipped to GAAP operating profit in FY2024 and expanded materially in FY2025. Net income remains negative because of interest expense on the convertible stack and unusual items, but the underlying operating engine has turned.
Cash flow & balance sheet
| Metric ($M) | FY2022 | FY2023 | FY2024 | FY2025 |
|---|---|---|---|---|
| Operating cash flow | (192) | (373) | 92 | 114 |
| Capex | (117) | (84) | (59) | (57) |
| Free cash flow | (309) | (456) | 33 | 57 |
| Cash & equivalents | 348 | 665 | 803 | 2,454 |
| Total debt | 1,021 | 1,455 | 1,530 | 2,992 |
| Net debt | 63 | 182 | 326 | 164 |
| Stockholders equity | 341 | 502 | 562 | 769 |
| Shares outstanding (M) | 205.7 | 224.7 | 229.1 | 280.0 |
The FY2025 balance sheet jumped because of the upsized $2.2B convertible senior notes priced late 2025, plus an October 2025 debt-for-equity exchange that retired ~$443M of old converts in exchange for ~$449M cash and 18.1M new shares. Cash position more than tripled, gross debt nearly doubled, and share count grew 22% in one year. Net debt fell despite the gross debt buildup because cash grew faster.
Capital structure detail: - $350M green convertible senior notes due 2029, 3.00% coupon (May 2024 issuance) - ~$2.2B convertible senior notes priced late 2025 - ~280M shares outstanding plus 3.53M Oracle warrants (struck $113.28, exp Oct 9 2026) - SK ecoplant preferred stock previously converted; SK still holds residual common stake
What is fueling growth today: 1. Oracle / AI data center pull — 1.2 GW already contracted under the master agreement, frame extends to 2.8 GW. This is the single biggest revenue driver in 2026–2027. 2. AEP utility procurement — 100 MW initial order placed, frame for up to 1 GW. Sets precedent for other regulated utilities to procure SOFC behind-the-meter. 3. Backlog buildout — reportedly $20B post-Oracle expansion vs. ~$6B at end of FY2025. Provides 4–5 years of revenue visibility at FY2027 consensus. 4. Capacity doubling — 2 GW annual nameplate by end of 2026 is the rate-limiter on revenue conversion. 5. Brookfield financing partnership — $5B AI deployment vehicle removes the customer-financing constraint for non-investment-grade buyers.
Pipeline / R&D: - R&D spend not separately broken out in headline release; historically ~6–8% of revenue. - Hydrogen electrolyzer line (Newark) at 2 GW capacity. Commercial scale-up pending policy and offtake clarity (US 45V tax credit dynamics, EU Green Deal pacing). - Marine application (Samsung Heavy pilot) — current materiality not verified. - M&A activity: none material disclosed in last 12 months; capital is being deployed into capacity not acquisitions.
| Contract / Program | Counterparty | Value | Status (April 2026) | What’s Left | Revenue Impact |
|---|---|---|---|---|---|
| Oracle Master Services Agreement | Oracle (ORCL) | Up to 2.8 GW; 1.2 GW already contracted; warrant for 3.53M shares at $113.28 | Expanded April 13, 2026; warrant issued April 9, 2026; deployment underway | Capacity build-out at Fremont/Newark; site permits at hyperscaler facilities | Material 2026 and 2027 revenue contributor; underpins the FY2027 consensus jump to $5.2B |
| AEP Frame Agreement | American Electric Power (AEP) | ~$2.65B over up to 1 GW | Frame agreement announced Nov 2024; 100 MW initial firm order Jan 2026 | Subsequent draw orders against the frame; state PSC approvals in service territories | First 100 MW ships through 2026; cadence of follow-on orders is the swing factor for FY2027–2028 |
| Brookfield Infrastructure Partnership | Brookfield Asset Mgmt (BAM) / Brookfield Infrastructure (BIP) | Up to $5B | Announced October 2025; structure is project-finance, not direct purchase | Specific project sites announced over time | Brookfield is the financing layer. Revenue still flows from end-customers (hyperscalers); Brookfield underwrites them. |
| SK ecoplant 500 MW Offtake (2023 expansion) | SK ecoplant | ~$1.5B product + ~$3B service over 20 years | Active, running through 2027 | Order cadence | Steady ~$200M/year product run-rate plus growing service tail |
| Equinix 19-Site Expansion | Equinix (EQIX) | >100 MW deployed across 19 IBX sites | Active, ~75 MW operational + ~30 MW under construction (Feb 2025) | Site commissioning | Reference customer; demonstrates uptime track record more than it drives the P&L |
The growth story is fundamentally contract-driven. If Oracle and AEP draw against their frames at the pace currently implied by consensus, FY2027 revenue lands at $5.2B and the stock can grow into its multiple. If draws slip by 12 months, the consensus path slips and the multiple looks indefensible.
| Risk | Likelihood | Existing Mitigants | Mgmt De-risk Plan | Can It Be Closed? |
|---|---|---|---|---|
| 1. Customer concentration (Oracle as new top customer) | High | Diversifying customer base (AEP, Brookfield, Equinix); Oracle warrant aligns interests | Brookfield financing partnership broadens customer pool to non-IG buyers; hyperscaler outreach beyond Oracle | Partly. Concentration shifts shape (from SK to Oracle) but cannot be eliminated at this growth rate. Will improve naturally as new hyperscalers convert. |
| 2. Capital-stack expansion / dilution | High | $2.45B cash buffer; FCF positive 2 years running; no near-term debt maturities | Self-funding signal in FY2026 guide; convertible structure delays dilution until conversion | Partly. Bloom can stop issuing new converts now that operations fund themselves. The Oracle warrant (3.53M shares) is the next dilution event before October 2026. |
| 3. CFO turnover (3 in 2 years) / governance | Medium | New CFO Edwards has Groq pedigree; CAO Kurzymski stayed through transition | Edwards onboarding Q2 2026; need clean Q1 print on April 28 | Yes if Edwards stays 24+ months and earnings calls show financial-reporting maturity. Closes by 2027 if no further turnover. |
| 4. Natural gas price exposure (customer ROI sensitivity) | Medium | Customers pay for time-to-power, not lowest-cost-of-energy | Hydrogen optionality; long-term contracts hedge customer side | No. Structural. Can only be hedged not eliminated. Mitigates as gas/power price spread widens. |
| 5. Service segment economics | Medium | Margin trajectory improving; Sridhar called out service margin gains on FY2025 call | Predictive maintenance on installed fleet; warranty redesign on newer SKUs | Yes if Service crosses to GAAP positive contribution by FY2026. Watch line-item breakout. |
| 6. Permitting / interconnect risk for AEP-style deals | Medium | Behind-the-meter siting avoids ISO interconnect; gas-line tie-in well-understood | State-by-state engagement; AEP partnership de-risks regulated utility path | Partly. State PSC processes still gate volume. |
| 7. Valuation / momentum risk | High | n/a (this is a market risk, not a company risk) | n/a | No. The multiple compresses on any execution miss; this is a structural feature of the current setup. |
Last earnings (Q4/FY2025, reported February 6, 2026): - Revenue: $777.7M for Q4 (+35.9% YoY); FY2025 $2.024B (+37.3%) - FY2025 operating income: $73M (vs. $23M FY2024) - Operating cash flow FY2025: $114M; FCF $57M - FY2026 guide: revenue $3.1–3.3B, non-GAAP operating income $425–475M, gross margin ~32% non-GAAP
Next earnings (Q1 2026): TODAY, April 28, 2026, after market close. - Consensus revenue: $540M (+66% YoY) per Yahoo aggregation of 19 analyst estimates - Consensus EPS: $0.13 (vs. $0.03 LY, +328%) - High estimate $674M / low $410M (wide dispersion)
Material news in last 90 days:
| Date | Event |
|---|---|
| Feb 6, 2026 | Q4/FY2025 print, FY2026 guide |
| Mar 2026 | Stock down ~13% on weak US jobs report; Jefferies maintained Underperform with $97 PT |
| Apr 9, 2026 | Oracle warrant issued (3.53M shares at $113.28, $400M strike value) |
| Apr 13–14, 2026 | Oracle frame expanded to 2.8 GW; stock +15% on session |
| Apr 14, 2026 | Jefferies upgraded Underperform → Hold, PT $97 → $187 |
| Apr 14, 2026 | JP Morgan raised PT $166 → $231 (Overweight) |
| Apr 21, 2026 | UBS raised PT $170 → $251 (Buy) — street high; thesis is “800 VDC data center revolution” |
| Apr 21, 2026 | Citigroup raised PT $162 → $229 (Neutral) |
| Apr 22, 2026 | Baird raised PT $172 → $242 (Outperform) |
| Apr 27, 2026 | Barclays raised PT $153 → $177 (Equal-Weight) |
| Apr 28, 2026 | Q1 2026 earnings (after close) |
The April analyst-revision wave is the key context for tonight’s print. Most desks have raised prints and PTs in the last two weeks specifically pricing in Oracle. Q1 needs to confirm the order intake is real.
Note on data: Pulled from yfinance (which aggregates 13F + Form 4 filings) as of latest reporting dates. Cross-check against latest DEF 14A and EDGAR Form 4s for material recent moves. SK ecoplant’s residual common-stock holding is filed as 13G/13D-type and shows up in insider roster as “Beneficial Owner of more than 10% of a Class of Security.”
Major holders breakdown: - Insider ownership: 5.82% - Institutional ownership: 86.0% (institutional float ownership: 91.3%) - Number of institutional holders: 1,106 - Short interest: 23.2M shares = 9.34% of float; 2.17 days to cover
| Holder | Type | Who They Are | Shares | % | Notable |
|---|---|---|---|---|---|
| Ameriprise Financial | Active asset manager | Wealth-management parent including Columbia Threadneedle’s Seligman Tech franchise | 28.65M | 10.14% | Largest active holder; Seligman Tech is a sector-specialist holding ~17M of these shares |
| BlackRock | Passive index | World’s largest asset manager; mostly index inclusion | 24.16M | 8.55% | +16.9% QoQ — index-driven buying as BE entered/grew weight in indices |
| Vanguard Group | Passive index | Same as BlackRock structurally | 20.85M | 7.38% | Stable position |
| Situational Awareness LP | Active sector specialist | Leopold Aschenbrenner’s AI infrastructure-focused fund (formed 2024) | 10.08M | 3.57% | New position; 100% increase in Q4 2025. The thesis-driven bet in the holder list. |
| Morgan Stanley | Multi-strategy | IB / asset mgmt holdings | 9.35M | 3.31% | +6.3% |
| D.E. Shaw | Quant | Systematic / multi-strategy hedge fund | 8.80M | 3.12% | -25.6% (trimming) |
| State Street | Passive index | SPDR / institutional indexer | 6.03M | 2.14% | +13.4% |
| Geode Capital | Passive index | Fidelity sub-advisor | 5.28M | 1.87% | +5.4% |
| Jane Street | Market maker / quant | Systematic liquidity provider | 4.83M | 1.71% | +71.9% (likely options-related hedging buildup) |
| Graticule Asia Macro Advisors | Macro | Adam Levinson’s Asia-focused macro fund | 4.76M | 1.69% | Stable |
Read on the holder mix: Passive (Vanguard, BlackRock, State Street, Geode) accounts for roughly 20% of shares outstanding, which is normal for an S&P-eligible name. The interesting holders are Ameriprise/Seligman (active conviction position) and Situational Awareness (thesis-driven AI infrastructure specialist). The 71.9% jump in Jane Street’s position likely reflects derivatives-related hedging activity around the post-Oracle option expansion, not a directional view.
| Insider | Position | Shares Held |
|---|---|---|
| KR Sridhar | CEO / Chair | ~4.72M (stock award Feb 27, 2026) |
| John T. Chambers | Director | 431K (5/14/2025 award) |
| Daniel Berenbaum | Former CFO | 300K (legacy holdings) |
| Shawn Soderberg | Officer | 517K (sold Apr 15 2026) |
| Jeffrey Immelt | Lead Independent Director | 230K (Mar 31 2026 award) |
| Satish Chitoori | COO | 212K (sold Apr 14 2026) |
| Aman Joshi | CCO | 181K (sold Apr 1 2026) |
| Mary K. Bush | Director | 134K |
| Michael Boskin | Director | 102K |
| Cynthia Warner | Director | (recent grant) |
| SK ecoplant Co., Ltd | 10%+ Beneficial Owner | 13.49M (most recent filing reflects July 2025 sale, not residual after sale) |
Recent insider activity (last 6 months, summary): 19 buys for 513K shares, 26 sells for 611K shares, net ~98K shares sold. April 2026 is concentrated selling at $135–$225 by Soderberg, Chitoori, Joshi after the Oracle pop. Sridhar himself acquired 300K via PSU vesting in February, has not been a recent open-market seller.
Activist filers: None active as a 13D in this session. SK ecoplant historically files as 13G strategic. Worth running an EDGAR 13D search before earnings if Pink wants confirmation.
| Metric | Value |
|---|---|
| Number of analysts | 24 |
| Buy / Hold / Sell breakdown | 13 Buy (incl. 4 Strong Buy) / 12 Hold / 2 Sell |
| Consensus rating | Buy / Moderate Buy |
| Mean PT | $166.96 |
| Median PT | $174.00 |
| High PT | $251 (UBS) |
| Low PT | $55 |
| Current price (Apr 27, 2026) | $234.68 |
The current price sits above the mean target and roughly at the 90th percentile of the analyst range. Two read-throughs: 1. The street has not fully caught up to the post-Oracle reality. Several recent revisions (Baird $242, UBS $251, Citi $229, JPM $231) cluster above $225, suggesting the mean is being dragged down by stale low-side targets. 2. Or the rally has overshot and the mean target is the right anchor.
The wide dispersion ($55 to $251) is itself diagnostic: this is a bimodal stock where bears think the multiple cannot hold and bulls think the AI infrastructure thesis is just getting started. Q1 earnings tonight is the first inflection-test of the bull case.
A full /filings BE pass should pull: - Q1 2026 10-Q
(after tonight’s print) - 10-K FY2024 detail on segment revenue split,
customer concentration disclosures, related-party SK ecoplant terms -
DEF 14A April 2025 for committee chairs, exec compensation, succession
provisions - Form 4s for Sridhar, Edwards (new CFO), and recent officer
sales - Any 8-K events in last 90 days (Oracle warrant filing was
8-K)
EDGAR PDF retrieval was blocked in this session via the standard fetcher; manual download or alternate fetch needed.
The headline question for this writeup is whether the people running Bloom Energy are aligned with shareholders, competent operators, and honest about what they sell to the street. The short answer is B-, improving. There are real flags, two of which deserve close watching: the CFO carousel (three CFOs in 24 months plus an 11-month interim) and the late-2025 capital allocation that monetized the existing convertible holders at the issuer’s expense. The biggest counterweight is that Sridhar’s personal stake remains roughly $750M and growing, the FY24-Q1 2026 guide history is a clean beat-and-raise pattern, and the recent Snabe board addition plus Edwards CFO hire are visible governance upgrades.
Background: BS Mechanical Engineering, NIT Tiruchirappalli (1982). MS Nuclear / PhD Mechanical, University of Illinois Urbana-Champaign (1989). NASA Ames researcher developing solid oxide electrolysis for Mars-mission O2 generation. Aerospace/Mechanical Engineering professor at University of Arizona, Director of the Space Technologies Laboratory. After NASA cancelled Mars Surveyor 2001, Sridhar inverted the architecture (run hydrocarbons through SOFC stack to generate electricity) and co-founded Ion America in 2001. Renamed Bloom Energy 2006. National Academy of Engineering inductee 2016.
Track record: Single-entity continuous founder. No prior failed companies. No public SEC enforcement action against him personally. No personal bankruptcy. The Bloom Box launched February 24, 2010 with the famous 60 Minutes feature; the company then went 8 years pre-IPO without commercial scale, raised >$1B private. The “20-year overnight success” arc obscures the fact that the unit economics did not work for most of that time. BE only crossed positive operating income for full year 2024.
Sources: Wikipedia – KR Sridhar; NASA Spinoff 2010; Fortune profile Apr 2026
Joined from Groq, where he was CFO and then briefly CEO. The Groq pedigree positions him as a narrative-aligned hire for the AI infrastructure rebrand, not for accounting depth. Groq is private and pre-revenue at scale, so Edwards has no public-company SOX battle scars. Whether that is an asset (fresh thinking) or a liability (no playbook for restatement landmines, especially given BE’s 2020 MSA history) is genuinely uncertain. His sign-on package details are in the April 13, 2026 8-K and the Form 3 filed April 21, 2026.
Owns the customer relationships including Oracle and AEP. Has not been CFO at any point (the original profile assumption was incorrect). Sold 10,000 shares at $135.88 on April 1, 2026 = $1.36M; current direct holdings 180,521 shares.
Manufacturing scale-up lead for the Fremont and Newark capacity ramps. Sold 20,000 shares at $204.23 on April 14, 2026 = $4.08M; current direct holdings 212,365 shares.
CAO since 2021. Stayed through the entire 11-month interim CFO period and continues as CAO under Edwards. The fact that he was kept on is a positive signal on transition continuity.
The headline forensic flag. Three CFOs in roughly two years plus an 11-month interim search at a company with a prior accounting restatement is governance noise that cannot be hand-waved.
| CFO | Tenure | Why They Left | Signal |
|---|---|---|---|
| Greg Cameron | Aug 2020 → Apr 29, 2024 | To CF Industries (June 17, 2024); 6-week gap suggests negotiated voluntary | Cameron was an Immelt hire (came from GE Capital where he was Immelt’s lieutenant). Departed right before FY24 op-income inflection. Probably voluntary, but the timing is odd — you don’t usually leave the moment the win is about to land. |
| Daniel Berenbaum | Apr 29, 2024 → May 1, 2025 | Twelve months exactly. Boilerplate Item 5.02(b) language (“amicable, not the result of any disagreement on accounting or financial policies”). No transition period. | This is the loudest specific signal. When a company doesn’t bother with a 60-90 day handoff, it usually means the relationship was broken. Most likely root cause: capital-policy disagreement preceding the $2.2B convert + Oracle warrant program (executed within 6 months of his exit). |
| Maciej Kurzymski (interim) | May 2025 → April 13, 2026 | 11-month interim is abnormally long. Typical CFO searches close in 4-6 months. | Either candidates kept turning down the role (red flag — high-profile CFOs read the same forensic signals you’re reading), or Sridhar was being picky in a market where his ideal candidate was unavailable. |
| Simon Edwards | April 13, 2026 → | From Groq | Narrative hire for AI infrastructure positioning. |
Sources: Cameron departure 8-K Apr 17 2024; CF Industries 8-K Jun 20 2024; Berenbaum exit – Seeking Alpha; Fuel Cells Works
| Name | Role | Shares | % of Out. | Est. Value at $288 | How Acquired |
|---|---|---|---|---|---|
| KR Sridhar | CEO/Chair | ~2,692,921 + 2.6M unvested | ~0.95% direct + 0.9% unvested | ~$775M direct + ~$750M unvested | Mix of founder shares, options exercises, RSU/PSU vesting |
| Aman Joshi | CCO | 180,521 | 0.06% | ~$52M | RSU/PSU + market sales |
| Satish Chitoori | COO | 212,365 | 0.07% | ~$61M | RSU/PSU + market sales |
| Shawn Soderberg | Officer | 517,463 | 0.18% | ~$149M | RSU/PSU + market sales |
| Jeffrey Immelt | Lead Independent Director | 229,741 | 0.08% | ~$66M | Director equity grants |
| John Chambers | Director | 431,157 | 0.15% | ~$124M | Director equity grants |
| Mary K. Bush | Director | 133,524 | 0.05% | ~$38M | Director equity grants |
| Michael Boskin | Director | 101,835 | 0.04% | ~$29M | Director equity grants |
Total insider ownership: 5.82%. All director holdings are stock awards, not open-market purchases. Sridhar himself has done 24 transactions over the past 5 years, all sells, zero open-market buys (GuruFocus). That said, the largest of those sales were forced by 10-year option expirations, not discretionary signaling.
| Date | Shares | Price | Value | Trigger |
|---|---|---|---|---|
| Mar 15, 2023 | 177,786 | ~$18 | $3.2M | Tax withholding on PSU vesting |
| Aug 25-29, 2025 | 216,955 | $48.97-$53.79 | $13.04M | Forced — exercising 10-year options struck $30.89 expiring Sep 10, 2025 |
| Feb 24, 2026 | 200,000 | $170-$170.05 | ~$34M | Discretionary — no options expiry trigger |
The Feb 2026 $34M sale was discretionary and represents ~7% of Sridhar’s personal stake. Not a flight, but the timing (after the stock 6x’d in 12 months) is rational diversification. What would be abnormal — a 10b5-1 plan filed right before bad news — is not present in the public record.
10b5-1 plans: Per the Q1 2026 filings, several officers (Soderberg, Chitoori, Joshi) appear to be selling under pre-existing 10b5-1 plans given the rhythm and pricing of the April 2026 sales. Specific 10b5-1 plan adoption dates not pulled in this pass.
| Name | Role | BE Stake (% of personal wealth, est.) | Other Public Co. Holdings | Private/Shell Interests | Wealth Concentration Read |
|---|---|---|---|---|---|
| Sridhar | CEO | Likely >80% of net worth in BE | None disclosed in DEF 14A | None disclosed | All-in. Strong conviction signal. |
| Immelt | Lead Indep. Director | Single-digit % | NEA portfolio companies (Desktop Metal, Tuya, Bright Health, Cleo, Collective Health, Formlabs, Sila Nano, TAE Technologies) + Twilio | NEA Venture Partner | Wealth concentration is at NEA, not BE. He is technically independent but functions as an NEA vehicle. |
| Chambers | Director | Single-digit % | JC2 Ventures portfolio (OpenGov, Pindrop, Quantum Metric, Sprinklr, Uniphore) | JC2 Ventures (his fund) | VC-on-board pattern; wealth at JC2. |
| Boskin | Director | Single-digit % | Multiple boards | Stanford Hoover affiliation | Academic; wealth distributed |
| Bush | Director | Single-digit % | Multiple corporate boards | Bush International LLC (consulting) | Distributed; consulting practice |
| Snabe | Director | <1% (joined Aug 2025) | Siemens AG Chairman | None disclosed | Wealth at Siemens; recent BE addition has not had time to accumulate |
Read: Sridhar’s wealth is concentrated in BE — strong alignment. The director group’s wealth is largely elsewhere (NEA portfolios, JC2 Ventures, Siemens), which is normal for outside directors but means the board’s personal financial conviction is moderate. Immelt as Lead Independent being an NEA partner is the specific concern: his economic decision-making weights NEA’s portfolio outcomes, not BE shareholders’ outcomes.
SK ecoplant Co., Ltd. is the four-vector related party that dominates this section:
This is the textbook configuration for usurpation-of-corporate-opportunity scrutiny. The Preferred Distributor Agreement (PDA) is not publicly filed in unredacted form. Pricing terms (most-favored-customer language, ASP discount %, take-or-pay structure) are not extractable from public disclosure. The 2024 10-K Note 17 discloses an SK-receivable balance of ~$93.5M and other related-party balances but does not give granular per-year revenue dollars by line item.
Econovation LLC is the SK ecoplant indirect-ownership vehicle. State of incorporation, registered agent, and other shell-detection metadata not pulled in this pass — recommended next step if Pink wants forensic depth here.
No IP licensing or consulting agreements between Bloom and Sridhar-controlled entities surfaced in this pass. Sridhar does not appear to have a personal LLC drawing fees from BE.
Bloom’s structure is relatively flat: parent corporation, Korean JV with SK, a couple of subsidiaries for international operations. No undercapitalized shell-asset holding entities surfaced.
2019-2020 Securities Class Action. Filed May 28, 2019 (N.D. Cal.). Allegations: false/misleading MSA accounting in IPO registration and through September 16, 2019. Bloom restated financials in February 2020 for periods Jan 2018 – Sept 2019. Settlement: $3.0 million — final approval granted by California federal judge. Lead counsel: Levi & Korsinsky / Kessler Topaz. (Stanford SCAC; Yahoo Finance coverage; Settlement site)
The $3M settlement is small relative to BE’s market cap and almost certainly insurance-funded. Material in signal value, not in dollars. The fact that the company has never publicly recommitted to specific accounting policy improvements with substance after the restatement is itself a transparency criticism.
No SEC enforcement action against BE or executives surfaces in public records. No personal litigation against Sridhar, Joshi, Chitoori, or Edwards.
| Component | Amount |
|---|---|
| Base salary | $876,923 |
| Bonus | $1,579,500 |
| Stock awards | $42,394,800 |
| Other compensation | $110,522 |
| Total compensation | $44,961,745 |
| Compensation Actually Paid (CAP) | $39,968,710 |
| Average non-PEO NEO comp | $3,755,269 (CAP $6,588,509) |
Peer group: Nasdaq Clean Edge Green Energy Total Return Index (an index, not a custom peer set). Using an index as the comp benchmark is a soft governance flag. It lets the comp committee avoid naming specific peers (PLUG, FCEL, STEM, etc.) and creates ambiguity in TSR comparisons. CEO pay ratio: not extracted.
Sources: Salary.com; DEF 14A 2025
Sridhar 2024 grants: - 1,500,000 PSUs + 500,000 RSUs = the regular package - 600,000 additional PSUs in two strategic grants (300K vested December 18, 2024 on disclosed criteria; 300K vesting before December 31, 2027 tied to “specific, objective criteria tied to specific strategic priorities” — strategic-priority criteria are not disclosed)
Performance metrics named in the proxy: Revenue (Product and Service), Non-GAAP gross margin, Non-GAAP operating income/loss, Cash flow from operations.
Maximum payout uplift: 200% (raised in the 2024 cycle from prior 150%). For the 2024 cohort, one-third of Sridhar’s 2024 equity package PSUs were achieved at 300% in year one. Translation: one tranche paid out at the maximum cap on year-one performance.
Hurdle vs. company’s own LT model. BE FY26 guide is $3.4-3.8B revenue, $600-750M op income (raised from $3.1-3.3B / $425-475M after Q1 2026). FY27 consensus $5.2B. PSU hurdles do not appear to be tied to those specific number thresholds — they are tied to broad metric categories with comp-committee discretion on percentage achievement. That is “pay for the easy beat” structure, not “pay for plan execution.”
| Hurdle Type | Target Specificity | Year-One Achievement | Company’s Own LT Plan | Read |
|---|---|---|---|---|
| Op income improvement | Broad category | 300% (max cap) on one tranche | Mgmt own guide implies +5x op income FY25→FY26 | Bar was set low relative to plan; max payout achieved easily |
| “Strategic priorities” tranche (300K PSUs) | Undisclosed | n/a | n/a | Most opaque of all grants; no public investor scrutiny possible |
| Other PSUs | Revenue/GM/OCF categories | Mixed | Plan implies massive growth | Generous structure with index peer benchmark |
Verdict: 200% max payout, plus 300% achievement on one tranche in year one, plus a 600K “strategic priorities” grant with undisclosed criteria, plus an index (not custom peer) benchmark. This is generous comp design with low transparency hurdles. It is not a “rip-off” — but it is the kind of structure that investors should haircut when valuing management quality.
| Date | Type | Amount | Strike/Price | BE Ref Price | Score |
|---|---|---|---|---|---|
| Jul 2018 | IPO | $270M | $15.00 | $15 | NEUTRAL (IPO price) |
| Sep 2018 | 2.50% Green Convert due 2025 | $290M | ~$13 conv | ~$10-13 | BAD (low) |
| 2020-21 | ATM equity / follow-on | $389M | ~$26 | $26+ | GOOD (above market) |
| Aug 2021 | 2.50% Green Convert exchange | — | — | — | NEUTRAL |
| Oct 2021 | SK ecoplant strategic preferred | $500M | ~$25 | ~$22-25 | GOOD |
| Sep 2023 | SK follow-on equity | $311M | ~$23 | ~$13-15 | GOOD (above market) |
| May 2024 | 3.00% Green Convert due 2029 | $402.5M (incl. greenshoe) | $20.85 conv | $15.73 | NEUTRAL (32.5% premium but stock was depressed) |
| Oct 2025 | Debt-for-equity exchange of 2028+2029 notes | ~$976M principal exchanged for ~42.4M shares + ~$988M cash | implied ~$23-24 effective | $127.85 | BAD — converted debt at deep-in-the-money strikes when stock had already 10x’d from issue |
| Nov 2025 | 0% Convertible due 2030 | $2.2B (upsized to $2.5B with greenshoe) | $194.97 conv | $127.85 | EXCELLENT — 52.5% premium, 0% coupon, peak optionality |
| Apr 2026 | Oracle warrant | up to $400M strike | $113.28 | $234 → $288 | MIXED — strike now ~60% below current; struck Oct 2025 when stock was ~$120; rational at the time but the warrant is now worth ~$620M of paper to Oracle at BE shareholder expense |
Sources: SEC 8-K Oct 30, 2025 debt exchange; Oracle CNBC Apr 13 2026
Apply the valuation-aware capital cost test. The True Equity Cost of Capital (TECC = 1 / forward P/E) tells you whether management is allocating capital rationally relative to actual cost.
| Year | Forward P/E (rough) | TECC | Action | Score |
|---|---|---|---|---|
| 2018-2021 | n/a (negative EPS) | n/a | Equity raises at $15-26 | Cannot judge against TECC; raised when capital window was open. Defensible. |
| 2022-2023 | n/a (negative EPS) | n/a | SK follow-on at $23 above market price | Good — they sold to a strategic above where the market was clearing. |
| May 2024 | n/a (still negative) | n/a | Convert at $20.85 with stock at $15.73 | Neutral. Convert is debt-like; doesn’t fail TECC test. |
| Nov 2025 | ~50x at $127 | ~2% | 0% convert at $194.97 (52.5% premium) | EXCELLENT. They sold optionality at TECC of ~2% with a 0% coupon. This is the textbook good action. |
| Oct 2025 | ~50x at $127 | ~2% | Debt-for-equity exchange forcing conversion at $23 effective strike | BAD. They retired debt that was already deep-in-the-money for 42.4M new shares — handing existing convert holders the optionality monetization at the issuer’s expense. This is the single worst-timed action. |
| Apr 2026 | ~75x at $234 | ~1.3% | Oracle warrant at $113.28 | Defensible — exchanged equity optionality for the Oracle volume commitment and PR halo, but warrant is now ~$620M ITM. Hard to grade until you see the realized revenue. |
Capital Allocation Timing Grade: NEUTRAL → GOOD, with one specific bad action in Oct 2025. The promotional-CEO exception applies partially — Sridhar has run a high-narrative arc that has kept the capital window open during years of cash burn, which is rational for a capital-dependent business. The Nov 2025 0% convert was the textbook good action. The Oct 2025 exchange was the textbook bad action. Net: C+, with B+ on the recent ones if you weight by amount.
No buyback program disclosed. Net issuer of shares throughout the company’s history. Share count went from 205.7M (FY2022) to 280.0M (FY2025) — a 36% increase in three years. This is not buyback discipline; it is sustained dilution to fund growth and convert holders.
| Year | Capex (M)|Revenue(M) | Incremental Rev / Capex | |
|---|---|---|---|
| FY22 | 117 | 1,201 | n/a baseline |
| FY23 | 84 | 1,335 | +$134M / -$33M = inverse — capex fell while revenue grew (mix shift) |
| FY24 | 59 | 1,474 | +$139M / -$25M = capex fell again (efficient growth) |
| FY25 | 57 | 2,024 | +$550M / -$2M = exceptionally capital-light growth in FY25 |
Capex efficiency grade: A. Bloom is generating large incremental revenue per capex dollar. This is partly mix (Product revenue scales without proportional capex once factories are built) and partly the fact that capacity expansion in 2025 was funded by service-revenue retention, not new capex.
Net rating accounts for: persistent dilution (negative), capital window timing on raises (mixed but improving), exceptional capex efficiency (positive), and the single worst-timed Oct 2025 exchange (negative).
The 2024-2026 guide-vs-actual pattern is clean beat-and-raise.
| Period | Initial Guide | Actual | Beat/Miss | Magnitude |
|---|---|---|---|---|
| FY2024 revenue | $1.6B (initial) | $1.474B | Miss | -8% (the one miss) |
| FY2025 revenue | $1.65B-1.85B (initial Feb 2025 guide) | $2.024B | Beat top of guide | +9% to +23% above range |
| FY2026 (initial) revenue | $3.1-3.3B (Feb 2026) | tracking $3.4-3.8B (raised Apr 28 post-Q1) | Raised on Q1 print | +10-15% above initial range |
| Q1 2026 revenue | implicit ~$540M consensus | $751.1M | Massive beat | +39% above consensus |
| Q1 2026 EPS | $0.13 consensus | not extracted in this pass | Beat | n/a |
The 2019-2023 era was the opposite — guide-and-cut cycle. The pivot to beat-and-raise coincides with the FY2024 op-income inflection and the new product-mix (AI data centers).
Verdict: Conservative / Sandbagger trending Straight Shooter in 2024-2026. They are guiding low and beating. This is what you want from a management team. Forward guidance is now a credible floor, not a ceiling.
Q1 2026 transcript (April 28, 2026, Motley Fool) shows Sridhar saying expansion to 2 GW capacity is “fully funded… funded from current resources over coming quarters.” This is a “no current plans to raise capital” statement delivered ~6 months after the $2.2B convert and Oracle warrant. Whether this commitment holds for the next 6 months is the watch-item.
Specific weasel-language patterns to watch: - Quarterly repetition of “fully funded” while issuing converts/warrants - Repetition of “AEP frame is firmed up” when only 100 MW has actually been ordered against the 1 GW frame - “Backlog growth from $6B to $20B” framing when $20B includes the Oracle frame at uncommitted-volume pricing
| Pattern | Frequency | Example | Read |
|---|---|---|---|
| “No current plans to raise capital” | Recurring | Q1 2026 call: “fully funded… current resources” (post $2.2B convert) | High frequency in fuel-cell sector; track quarterly |
| “On track for X” (where X is product launch / capacity / certification) | Moderate | 2 GW capacity by end-2026 | Mostly accurate so far |
| “Subject to” | Light | Used around AEP regulatory approvals | Normal qualifying language |
| “We expect operating leverage at scale” | Moderate (older calls) | Pre-2024 era | Now actually delivering |
| “Strong pipeline” | Heavy | Used quarterly | Hard to verify; backlog disclosure has gotten more specific |
Verdict: Moderate weasel-language frequency. Below PLUG levels (which are extreme) but not as clean as a top-tier industrial. The shift to beat-and-raise has reduced weasel reliance.
Overall Follow-Through Rate (estimated, last 8 quarters): ~75-85% Guidance Tendency: Conservative trending Straight Shooter Weasel Language Frequency: Low-Moderate
Credibility band: Mid-to-High. Discount forward statements modestly but treat the current beat-and-raise pattern as real. The transition coincides with the CFO turnover, so watch whether Edwards continues the conservative posture. If he loosens guidance discipline, the credibility grade should drop.
10-member board, 7 outside directors, 2 inside (Sridhar + 1 other if any inside director — verify in latest proxy). Lead Independent Director structure.
| Director | Tenure | Other Boards / Affiliations | Independence Assessment |
|---|---|---|---|
| KR Sridhar | Founder | n/a | Not independent (combined Chair/CEO) |
| Jeffrey Immelt (Lead Indep.) | Since May 2020 | NEA Venture Partner; Twilio; NEA portfolio companies (Desktop Metal, Tuya, Bright Health, Cleo, Collective Health, Formlabs, Sila Nano, TAE Technologies) | Quasi-independent. NEA was a pre-IPO Bloom investor. Immelt is technically independent but functions as an NEA vehicle. Hired Greg Cameron (the first of the three CFO carousel members). |
| Jim Snabe | Since Aug 2025 | Chairman Siemens AG, ex-co-CEO SAP, Allianz SE board | Genuinely independent. Strong recent addition. |
| John Chambers | Since May 2018 | JC2 Ventures (his fund); OpenGov, Pindrop, Quantum Metric, Sprinklr, Uniphore | Independent (VC pattern; no BE-specific conflict) |
| Michael Boskin | Long-tenured | Stanford Hoover; multiple boards | Independent but long-tenured (entrenchment risk) |
| Mary K. Bush | Long-tenured | Bush International LLC; multiple boards | Independent |
| Cynthia Warner | Since June 2023 | ex-CEO Renewable Energy Group; prior BP exec | Independent; energy-relevant |
| Eddy Zervigon | Pre-IPO | Quantum Xchange CEO; Riverside Management Group Special Advisor | Independent but long-tenured (Chair of Nominating Committee) |
5 of 7 outside directors are genuinely independent. Immelt’s NEA affiliation and Zervigon’s pre-IPO tenure are the two soft flags.
Specific committee chairs not extracted in this pass. Critical to verify the Audit Committee has at least one designated financial expert with restatement-experience pedigree given the 2020 MSA history. Recommended next step: pull DEF 14A 2025 committee section.
Not pulled in this pass. Worth checking if any compensation-related proposals have failed (would indicate captured comp committee).
| Dimension | Rating | Key Finding |
|---|---|---|
| Skin in the Game | 🟢 Green | Sridhar ~$775M direct holdings; founder concentration |
| Holdings Concentration | 🟡 Yellow | Sridhar all-in; director group wealth elsewhere (Immelt at NEA) |
| Shell / Cross-Holdings | 🟡 Yellow | SK ecoplant 4-vector related party; PDA terms not public |
| Capital Allocation | 🟡 Yellow | Mixed timing record; Oct 2025 exchange = bad; Nov 2025 0% convert = excellent |
| Compensation Alignment | 🟡 Yellow | 200% max + 300% paid in year 1; “strategic priorities” undisclosed; index peer benchmark |
| Credibility / Follow-Through | 🟢 Green | Beat-and-raise 2024-Q1 2026; conservative guidance |
| Governance Quality | 🟡 Yellow | Combined Chair/CEO; Lead Indep. is NEA partner; Snabe/Edwards = upgrades |
| Litigation / Enforcement | 🟡 Yellow | 2020 restatement + $3M class action; no SEC enforcement |
| Overall Management Grade | B- | Improving from C; still has 4 yellow flags worth monitoring |
None at the deal-breaker tier.
These are not bad operators. Sridhar has built one of two companies in the world that can deliver gigawatt-scale on-site power for AI data centers within timeframes that hyperscalers need, and the 2024-2026 inflection in guidance discipline and operating margins is real. The CFO carousel is a real governance flag but does not appear to mask underlying accounting fraud. The compensation structure is generous and lightly transparent, but Sridhar’s personal stake is large enough that his economic interests are aligned with shareholders.
The watch-items if you own this: (i) does Edwards continue the conservative guidance pattern or loosen it; (ii) does the company stop issuing dilutive paper now that operations fund themselves; (iii) does the audit committee disclose tightened controls in the next 10-K following the 2020 restatement legacy; (iv) does any further related-party expansion with SK ecoplant get fairness-opinion oversight.
Would I trust them with my capital? Yes, with a one-multiple-turn governance discount baked into valuation. This is a B- governance grade, not an A. The stock is priced for execution; if execution slips, the multiple compresses fast and the governance flags get amplified by the market.
| Scenario | FY35 Revenue | Terminal Op Margin | WACC | Terminal Growth | Implied Fair Value | vs. $287.97 |
|---|---|---|---|---|---|---|
| Bear | $15B | 18% | 11% | 3% | ~$72 | -75% |
| Base | $23B | 25% | 10% | 4% | ~$136 | -53% |
| Bull | $35B | 30% | 8% | 5% | ~$295 | +2% |
The base case fair value is roughly half the current price. The current stock price requires either bull-case revenue scaling AND bull-case margins AND a sub-9% discount rate. Each of those is independently defensible. Together, they require essentially flawless execution against the AI data-center power thesis for a decade. The setup is asymmetric: small misses on any of the three legs compress the multiple meaningfully; large beats are needed to expand it further from here.
This is the textbook valuation problem for a momentum stock. DCF math doesn’t drive the price; the AI infrastructure scarcity narrative does. The model is useful for quantifying what scenarios are baked in, not for signal-generation on entry/exit timing.
BE’s FY2024 revenue ($1,474M) breaks into four segments. FY2025 $2,024M segment split not yet broken out at line-item level in the headline release; using FY2024 mix as the base.
| Segment | FY2024 ($M) | % | FY26E mid | FY30E | FY35E | Notes |
|---|---|---|---|---|---|---|
| Product | 1,090 | 74% | 2,500 | 8,400 | 16,500 | Energy Server hardware. Volume scales linearly with capacity expansion. Oracle 2.8 GW + AEP 1 GW = primary drivers. |
| Service | ~210 | 14% | 350 | 1,200 | 3,200 | 15-year fixed-fee O&M contracts attached to each Product sale. Recurring, growing tail. Margin recovery from negative to ~30% over the model period. |
| Installation | ~150 | 10% | 600 | 1,800 | 2,500 | Site engineering. Grew 31.8% YoY in FY24. Lower-margin pass-through. |
| Electricity | ~25 | 2% | 150 | 600 | 800 | Managed-services / PPA. Long-dated. |
| Total | 1,474 | 100% | 3,600 | 12,000 | 23,000 | Mgmt FY26 guide $3.4-$3.8B post Q1 2026 raise |
Key revenue assumption: Product revenue compounds at ~28% CAGR through 2030 driven by Oracle/AEP/Brookfield contract drawdowns, then decelerates to mid-teens through 2035 as the AI buildout cycle matures. Service revenue grows as the installed base compounds (each new Product sale adds 15+ years of Service tail). Installation tracks Product. Electricity remains the smallest line.
FY2026 explicit: $3.6B (mgmt guide midpoint of raised range $3.4-$3.8B) FY2027 explicit: $5.24B (current consensus average) FY2028 explicit: $7.2B (+38%) FY2029 explicit: $9.5B (+32%) FY2030 explicit: $12.0B (+26%) FY2031-2035: declining growth from 21% to 7% as cycle matures
| Year | Revenue | Op Margin | Op Income | Notes |
|---|---|---|---|---|
| FY2022 (actual) | 1,201 | -22% | -261 | Pre-inflection era |
| FY2023 (actual) | 1,335 | -16% | -209 | Approaching breakeven |
| FY2024 (actual) | 1,474 | +1.6% | +23 | The inflection |
| FY2025 (actual) | 2,024 | +3.6% | +73 | Expansion |
| FY2026E | 3,600 | 19% | 685 | Mgmt guide $600-750M = ~17.6%-19.7% midpoint |
| FY2027E | 5,237 | 22% | 1,152 | |
| FY2028E | 7,200 | 24% | 1,728 | |
| FY2029E | 9,500 | 25% | 2,375 | Terminal margin reached |
| FY2030E | 12,000 | 25% | 3,000 | |
| FY2031E | 14,500 | 25% | 3,625 | |
| FY2032E | 17,000 | 25% | 4,250 | |
| FY2033E | 19,500 | 25% | 4,875 | |
| FY2034E | 21,500 | 25% | 5,375 | |
| FY2035E | 23,000 | 25% | 5,750 |
Margin justification: Terminal 25% op margin sits between hardware peers (Cummins ~13-15%, GE Vernova ~10-12%) and software/IP-heavy industrial peers (Watts Water ~18%, Roper ~30%). Bloom’s mix justifies the premium because: (i) Service is high-incremental-margin once installed-base scales; (ii) Product is increasingly software-rich (digital twin, predictive maintenance); (iii) hyperscaler customer concentration enables fixed-price contracts at favorable terms during the AI build cycle. The 25% terminal is closer to a base-bear blend; 30% would be the bull case.
Risk to terminal margin: AI data center demand intensity declines after 2030 → Service margin holds but Product mix shifts toward more competitive utility procurements at lower margin. Pricing pressure from Korean/Chinese SOFC entrants by 2030. Hydrogen transition adds capex without immediate revenue.
This is the methodologically interesting part. Two cost-of-equity frameworks give very different numbers.
| Component | Value |
|---|---|
| Risk-free rate (10Y UST) | 4.3% |
| Equity risk premium | 5.0% |
| Industry-adjusted beta | 1.5 (Yahoo raw beta is 3.19; the high beta reflects momentum-volatility, not fundamental risk) |
| Cost of equity | 11.8% |
| Pre-tax cost of debt | 3.0% (BE’s recent 0% convert + 3% green convert) |
| After-tax cost of debt | 2.4% |
| Capital structure (market values) | ~95% equity / 5% debt |
| WACC (CAPM) | 11.3% |
TECC = 1 / forward P/E.
At $287.97 and forward EPS $4.02, forward P/E = 71.6x. TECC ≈ 1.4%.
The market is currently pricing BE at a cost of equity of ~1.4%, which is structurally indefensible long-term but reflects the AI infrastructure scarcity premium. The gap between CAPM-WACC (11.3%) and TECC (1.4%) is the 9.9-point momentum premium that the market is awarding to “pick-and-shovel AI” stories right now.
WACC = 10% (round-number sensible mid-point that gives roughly 5-year-Treasury-plus-equity-premium math while not crediting the full TECC compression). Sensitivity table runs WACC 8% / 10% / 12% to show the bull/base/bear envelope.
| Item | FY26E | FY30E | FY35E (terminal) |
|---|---|---|---|
| D&A as % of revenue | 2.5% | 4.0% | 4.0% |
| Capex as % of revenue | 4.0% | 5.0% | 5.0% |
| Working capital as % of ΔRevenue | 10% | 10% | 10% |
| Net reinvestment rate (capex+WC-D&A) / NOPAT | ~30% | ~25% | 16% (terminal) |
Capex justification: Bloom is currently capital-light because Fremont and Newark capacity additions are largely complete. Capex/revenue runs ~3% in 2025. Once capacity hits 2 GW (end 2026), growth requires another factory or expansion. Long-run capex/revenue 5% is reasonable for a hardware-plus-service business at scale.
Working capital: Bloom historically runs material WC drag during ramp (inventory + receivables). 10% of ΔRevenue is consistent with industrials at this growth rate.
Use the implied-reinvestment formulation: TV = NOPAT_t+1 × (1 - g/ROIC) / (WACC - g)
| Input | Value |
|---|---|
| Terminal NOPAT (FY36, 4% growth from FY35) | $4,720M |
| Terminal ROIC | 25% |
| Terminal growth (g) | 4% |
| Reinvestment rate (g/ROIC) | 16% |
| Terminal FCF (NOPAT × (1 - 16%)) | $3,965M |
| WACC - g | 6% |
| Terminal value (end of year 10) | ~$66B |
The 4% terminal growth is on the high side of conventional modelers’ bands (which typically cap at 2.5-3.5%) but reasonable for a structurally-tailwinded business. AI compute demand will plausibly grow faster than nominal GDP for the next 15-20 years; even after AI growth normalizes, the on-site power TAM is structurally tied to electrification trends.
| Year | Revenue | Op Inc | NOPAT | D&A | Capex | ΔWC | UFCF | Discount factor | PV |
|---|---|---|---|---|---|---|---|---|---|
| FY26 | 3,600 | 685 | 541 | 90 | 144 | 158 | 329 | 0.909 | 299 |
| FY27 | 5,237 | 1,152 | 910 | 157 | 236 | 164 | 667 | 0.826 | 551 |
| FY28 | 7,200 | 1,728 | 1,365 | 252 | 360 | 196 | 1,061 | 0.751 | 797 |
| FY29 | 9,500 | 2,375 | 1,876 | 380 | 475 | 230 | 1,551 | 0.683 | 1,059 |
| FY30 | 12,000 | 3,000 | 2,370 | 480 | 600 | 250 | 2,000 | 0.621 | 1,242 |
| FY31 | 14,500 | 3,625 | 2,864 | 580 | 725 | 250 | 2,469 | 0.564 | 1,394 |
| FY32 | 17,000 | 4,250 | 3,358 | 680 | 850 | 250 | 2,938 | 0.513 | 1,508 |
| FY33 | 19,500 | 4,875 | 3,851 | 780 | 975 | 250 | 3,406 | 0.467 | 1,589 |
| FY34 | 21,500 | 5,375 | 4,246 | 860 | 1,075 | 200 | 3,831 | 0.424 | 1,624 |
| FY35 | 23,000 | 5,750 | 4,543 | 920 | 1,150 | 150 | 4,163 | 0.386 | 1,605 |
| PV of explicit | 11,667 | ||||||||
| Terminal value | 0.386 | 25,476 | |||||||
| Enterprise value | 37,143 | ||||||||
| Less: net debt | (164) | ||||||||
| Plus: cash optionality (Oracle warrant, partly offset by dilution) | (300) | ||||||||
| Equity value | 36,679 | ||||||||
| Diluted shares (post-Oracle warrant exercise) | 290M | ||||||||
| Implied fair value per share | ~$126 |
(All $ in millions except per-share)
Adjustments to think about: - Oracle warrant intrinsic value of ~$620M at current price is a transfer to Oracle; net dilution to existing holders embedded - Convertible note overhang (~$2.55B principal) needs to convert at strike ~$195 — already in the money; conversion adds ~12.5M shares = 4.4% additional dilution beyond the 290M base - NOLs from accumulated losses give effective tax shield in early years (model uses 21% throughout for conservatism)
Range with adjustments: $125-$140 base-case fair value per share.
At $287.97 × 290M diluted = $83.5B equity, $83.7B EV.
Backing out the implicit terminal value: - PV of explicit (10 years) at base assumptions = $11.7B - Implied PV of terminal = $83.7B - $11.7B = $72B - Implied terminal value (year 10) = $72B / 0.386 = $186B - At 6% (WACC - g) and 84% FCF/NOPAT conversion, implied terminal NOPAT = $186B × 0.06 / 0.84 = $13.3B - Backing into FY35 NOPAT = $13.3B / 1.04 = $12.8B - Implied FY35 op income = $12.8B / (1 - 21%) = $16.2B
To justify $288, the model needs FY35 op income of $16.2B — roughly 3x the base-case $5.75B.
That requires one of: 1. FY35 revenue of $65B at 25% margin (vs. base case $23B) 2. FY35 revenue of $46B at 35% margin (impossible margin) 3. A combination plus lower discount rate (8%)
The first option requires BE to be doing $65B in revenue by 2035 — that is roughly today’s Cummins revenue scale. It implies that on-site fuel cells become a 25% market share solution to ~30% of the AI data center power buildout. That’s not a fantasy, but it requires the Oracle/AEP/Brookfield triangle to be replicated 10x over by other hyperscalers.
The third option (lower WACC) is what the market is actually doing — accepting TECC of 1.4% as the implicit cost of capital. That works as long as the AI scarcity narrative holds. When it cracks, the multiple compresses by half in weeks.
Sensitivity of fair value per share ($) to terminal op margin and WACC (base case revenue path):
| Terminal margin / WACC | 8% | 9% | 10% | 11% | 12% |
|---|---|---|---|---|---|
| 18% | $175 | $138 | $115 | $94 | $76 |
| 22% | $230 | $185 | $150 | $122 | $100 |
| 25% (base) | $275 | $215 | $175 | $145 | $115 |
| 28% | $325 | $255 | $205 | $170 | $140 |
| 32% (bull) | $385 | $305 | $240 | $200 | $165 |
(Values rounded; back-of-envelope reproducibility ±5%)
Read: Current price $288 is consistent with terminal margin of 28-32% and WACC 8-9%. Both ends of that band are bullish but not crazy. The price is defensible if you believe the AI data center power TAM is large enough to sustain ~28%+ operating margins for a decade and that the appropriate discount rate is closer to a high-quality industrial than a typical small-cap growth name.
Sensitivity to revenue path (terminal margin 25%, WACC 10%):
| FY30 Revenue / Terminal growth | 3% | 4% (base) | 5% |
|---|---|---|---|
| $10B (slower ramp) | $98 | $115 | $138 |
| $12B (base) | $115 | $135 | $162 |
| $15B (faster ramp) | $138 | $162 | $195 |
| $18B (bull) | $158 | $185 | $222 |
Probability-weighted blend across three scenarios:
| Scenario | Probability | Fair value | Probability-weighted |
|---|---|---|---|
| Bull (Oracle deal replicates 5x; FY30 rev $18B; terminal margin 30%; WACC 8%) | 25% | $295 | $74 |
| Base (current trajectory holds; FY30 rev $12B; terminal margin 25%; WACC 10%) | 50% | $136 | $68 |
| Bear (one major hyperscaler walks; backlog conversion slips 18 months; terminal margin 18%; WACC 11%) | 25% | $72 | $18 |
| Probability-weighted fair value | $160 |
Probability-weighted base ~$160 vs. current $288 = ~45% downside to weighted-fair. Even the 25%-weighted bull case ($295) only roughly matches current price. The asymmetry is unfavorable from here.
What would change this picture: - Bull weight increases if a second hyperscaler signs a 1+ GW frame in next 6 months - Bear weight increases if AEP draws against the 1 GW frame slip beyond initial 100 MW, or if the Korean/Chinese SOFC competition materializes faster than expected
The DCF cannot justify the current price under base-case assumptions. The price requires bull-case revenue, bull-case margins, and bull-case discount rate simultaneously. Probability-weighted fair value sits ~45% below the current quote.
That does not automatically mean “sell.” Momentum stocks can sustain price-to-DCF gaps for years if the narrative compounds. The reverse-DCF tells you what the market is buying: a 10-year story where Bloom becomes a $50-65B revenue industrial supplier of choice for AI data center power, and where the market continues to accept TECC of 1.4% during that buildout.
Right way to use this DCF: 1. Scenario weighting is the actual decision. If your subjective probability that BE replicates the Oracle deal 3-5 times is >40%, the price makes sense. Below 25%, the DCF math says exit. 2. Watch for the catalysts that move the bull weight — second hyperscaler frame, accelerated AEP draw, Brookfield project pipeline announcements. 3. The implied multiple compression on a missed catalyst is severe. A single quarter where backlog growth flatlines or AEP draw slips would compress the multiple from 75x forward EPS to 40-50x — call it a 30-40% correction.
Position-sizing implication: This is a stock for a small-to-medium position size with a defined exit discipline, not a core long-term hold at this multiple. The asymmetric payoff requires you to manage the position actively.
| Item | Amount | Notes |
|---|---|---|
| Cash & equivalents (FY25) | $2,454M | Funded by Nov 2025 $2.5B convert (upsized) |
| Short-term debt | $378M | |
| Long-term debt | $2,614M | Includes 2029 and 2030 converts |
| Total debt | $2,992M | |
| Net debt | $164M | Cash buffer covers gross debt almost completely |
| Convertible overhang (strike-weighted) | ~$2.55B principal | 2029 notes (3% coupon, $20.85 strike) + 2030 notes (0% coupon, $194.97 strike). Both ITM at $288 |
| Oracle warrant | 3.53M shares at $113.28 strike | Expires Oct 9, 2026; ~$620M intrinsic value at current price |
| Implied conversion dilution | ~12.5M shares (~4.4%) | Beyond the 280M FY25 base |