BE
Bloom Energy Corporation
⚡ AI Power
$277.00
Mkt Cap ~$79.9B
52W: $17.01 – $302.99 1
Next earnings: 30 Jul 2026
👁 Watch
Entry zone: $180–$220
Do not add above $230
The market has correctly identified Bloom's behind-the-meter fuel cell as the fastest solution to AI data center power constraints — but the Q1 2026 beat and $7.65B contract haul have already priced a multi-year execution scenario; at 125–135× non-GAAP fwd P/E, there is no margin of safety for new entry and the fundamental DCF puts fair value at $57–$160, not $277.
TTM P/E (NG)
~145×
⚠ recently profitable
Fwd P/E (NG)
~135×
PEG 3.37
Rev YoY (Q1)
+130%
$751M vs $326M
Gross Margin
30.0%
31.5% non-GAAP
OCF Margin
~10%
Q1 annualised
Thesis Durability
B+
Variant perception: Grid interconnection backlogs in VA, TX, and OR run 3–7 years; Bloom's SOFC deploys in 55 days behind the meter — that speed gap is structural for this decade, not a cyclical blip.
TAM: Wells Fargo projects US AI power demand up 550% by 2026 to 52 TWh, then +1,150% to 652 TWh by 2030.2 Bloom's $7.65B in contracts (Oracle 2.8 GW, AEP $2.65B 20-year, Equinix 100 MW+, Brookfield $5B financing) in a 90-day window is demonstrably real revenue, not narrative.3
Moat: Long-term service agreements create switching cost (Bloom owns the stack maintenance); Oracle's $400M warrant creates hyperscaler alignment; ceramic SOFC stack has 20+ years of IP development — not easily replicated at scale by GE Vernova or battery alternatives on the current 2026 timeline.
Bear case: The moat window is finite. IRA Section 48E clean electricity ITC (30% credit) is politically exposed — Bloom's own 10-K names "One Big Beautiful Bill" modifications as a material risk.4 Natural gas dependency introduces commodity price risk vs. solar+battery. Grid expansion (FERC transmission reform, MISO/PJM capacity auctions) could narrow the deployment speed advantage by 2028.
TAM: Wells Fargo projects US AI power demand up 550% by 2026 to 52 TWh, then +1,150% to 652 TWh by 2030.2 Bloom's $7.65B in contracts (Oracle 2.8 GW, AEP $2.65B 20-year, Equinix 100 MW+, Brookfield $5B financing) in a 90-day window is demonstrably real revenue, not narrative.3
Moat: Long-term service agreements create switching cost (Bloom owns the stack maintenance); Oracle's $400M warrant creates hyperscaler alignment; ceramic SOFC stack has 20+ years of IP development — not easily replicated at scale by GE Vernova or battery alternatives on the current 2026 timeline.
Bear case: The moat window is finite. IRA Section 48E clean electricity ITC (30% credit) is politically exposed — Bloom's own 10-K names "One Big Beautiful Bill" modifications as a material risk.4 Natural gas dependency introduces commodity price risk vs. solar+battery. Grid expansion (FERC transmission reform, MISO/PJM capacity auctions) could narrow the deployment speed advantage by 2028.
Sound but time-bounded. Real TAM, real contracts, real deployment advantage — but the moat window is 3–5 years, not 10+. Not structurally defensible like NVDA CUDA or TSM node leadership.
Business Quality
B
Q1 2026 inflection: Revenue +130.4% YoY ($751M), first GAAP profit in Q1 ever ($70.7M net income), OCF positive for first time in Q1 ($73.6M).5 Adj. EBITDA $143M vs $25.2M — 6× improvement.
Margin trajectory: Non-GAAP gross margin 31.5% (FY2026 guided 34%). Operating margin 17.3% (up 1,300 bps YoY). EPS estimates revised sharply upward post-Q1.
Balance sheet: $2.52B cash, $2.60B recourse debt (net ~$80M recourse); $2.2B of that is 0% convertible notes due 2030.6 Manageable, but convertible dilution risk if stock trades near or below conversion price on refinancing.
CFO transition: Simon Edwards appointed CFO on 26 Mar 2026 — mid-execution CFO change during the most complex manufacturing ramp in company history is a mild governance flag.7
Form 4 (last 90 days): Multiple executives selling under 10b5-1 plans adopted 26 Nov 2025 (pre-Oracle deal): CCO Joshi sold 10,000 shares at $175.60 (Feb); CFO Soderberg sold 42,881 shares at $149–161 (Mar) and 55,000 more at $204–225 (Apr); Director Mary Bush sold shares.8 Plans were pre-set before the major catalysts, but breadth and volume across multiple executives is worth monitoring.
Margin trajectory: Non-GAAP gross margin 31.5% (FY2026 guided 34%). Operating margin 17.3% (up 1,300 bps YoY). EPS estimates revised sharply upward post-Q1.
Balance sheet: $2.52B cash, $2.60B recourse debt (net ~$80M recourse); $2.2B of that is 0% convertible notes due 2030.6 Manageable, but convertible dilution risk if stock trades near or below conversion price on refinancing.
CFO transition: Simon Edwards appointed CFO on 26 Mar 2026 — mid-execution CFO change during the most complex manufacturing ramp in company history is a mild governance flag.7
Form 4 (last 90 days): Multiple executives selling under 10b5-1 plans adopted 26 Nov 2025 (pre-Oracle deal): CCO Joshi sold 10,000 shares at $175.60 (Feb); CFO Soderberg sold 42,881 shares at $149–161 (Mar) and 55,000 more at $204–225 (Apr); Director Mary Bush sold shares.8 Plans were pre-set before the major catalysts, but breadth and volume across multiple executives is worth monitoring.
Solid with caveats. Clear inflection in OCF and GAAP profitability; but 30–34% gross margins are low for an AI-narrative multiple, SBC gap ($0.25 GAAP vs $0.44 non-GAAP EPS) is real, and systematic insider selling on pre-set plans continues at volume.
Entry Price Discipline
D
Fwd P/E: ~135× (at $277 ÷ $2.05 guidance midpoint EPS). GuruFocus NTM consensus: 121.69×.9 Seeking Alpha non-GAAP fwd: 187×. Range 121–187×.
vs sector: AI Power sector — GEV trades at ~55× fwd P/E (the most growth-rated AI power name); NEE at ~22×. BE at 121–135× is 2.2–2.5× the sector maximum — firmly D per grading rubric (>2× sector median).
PEG: 3.37 — well above the 1.0–2.0 range where premium is defensible for a capital-intensive hardware manufacturer.
Distance from entry zone: Current $277 is 26–54% above my entry zone of $180–$220. At $200, BE trades at ~97× on $2.05 EPS — still expensive, but partially justified by 80% revenue growth and expanding margins.
Q1 beat: Revenue $751M vs $539M est (+39% beat); EPS $0.44 vs $0.13 est (+238% beat). The beat is now embedded in the multiple.5
My EPS variant view: FY2026 guidance ($1.85–$2.25) is credible given Q1's momentum, but FY2027 upside requires Oracle 2.8 GW to deploy on schedule AND gross margins to reach 38–40%. Product mix, manufacturing overhead, and installation losses (−35% margin in Q1) are structural headwinds to rapid margin expansion.
vs sector: AI Power sector — GEV trades at ~55× fwd P/E (the most growth-rated AI power name); NEE at ~22×. BE at 121–135× is 2.2–2.5× the sector maximum — firmly D per grading rubric (>2× sector median).
PEG: 3.37 — well above the 1.0–2.0 range where premium is defensible for a capital-intensive hardware manufacturer.
Distance from entry zone: Current $277 is 26–54% above my entry zone of $180–$220. At $200, BE trades at ~97× on $2.05 EPS — still expensive, but partially justified by 80% revenue growth and expanding margins.
Q1 beat: Revenue $751M vs $539M est (+39% beat); EPS $0.44 vs $0.13 est (+238% beat). The beat is now embedded in the multiple.5
My EPS variant view: FY2026 guidance ($1.85–$2.25) is credible given Q1's momentum, but FY2027 upside requires Oracle 2.8 GW to deploy on schedule AND gross margins to reach 38–40%. Product mix, manufacturing overhead, and installation losses (−35% margin in Q1) are structural headwinds to rapid margin expansion.
⚠ Short interest data: no figure found from Fintel free tier at time of research — not verified.
Priced to perfection and beyond. The stock reflects the bull scenario in full. Fundamental DCF implies fair value $57–$160; current $277 offers no margin of safety regardless of scenario.
Verdict — Watch · Do Not Add Above $230
Bloom Energy's AI power thesis has made the transition from narrative to demonstrated execution — $7.65B in 90-day contracts, Q1 revenue +130%, first GAAP Q1 profit in company history, and Oracle deploying 1.2 GW imminently are not hypothetical. The thesis is real. The problem is valuation: the market has already repriced this into the stock, driving it from a $17 low in April 2025 to $277 today (+1,529%). At ~135× non-GAAP fwd P/E with a PEG of 3.37, there is no margin of safety — the bull case on a fundamental DCF supports ~$160, not $277. Early entry in April 2025 at ~$17 provides a cost-basis floor that new entry near ATH does not share — the risk profile is fundamentally different. The pullback thesis is credible (beta 3.2, 52W high $302.99, stock still −8% below ATH) but Watch — not Add — is the correct stance until a pullback to $180–$220. The entry zone becomes a Strong Add if Q1 2026 execution continues to print in Q2 (guided "at least as good as Q1") while the stock retreats on macro noise.
New position: Do not initiate above $230. Entry zone $180–$220 → size to 2–3% of total portfolio (starter/watching conviction tier). If Q2 2026 print confirms execution ($750M+ revenue, margins holding), re-evaluate entry zone and potential scale to 5% at $220 or below. Max 20% single-stock cap — this name does not merit high conviction sizing at current price.
Opportunity-cost check
▾
vs CSPX
CSPX owns every major AI infrastructure beneficiary with diversified exposure and ~17× P/E; BE at 135× fwd P/E on a hardware company with 30% gross margins offers no risk-adjusted edge over CSPX from this price. The edge only re-emerges below $220 where a 90–100× multiple on a high-growth AI power thesis becomes a defensible premium.
vs current holdings
BE introduces new "AI Power" factor exposure not in CSPX, IBIT, MU, NOW, PLTR, META, or GOOGL — no direct displacement. The relevant comparison is opportunity cost vs adding to MU (fwd P/E ~7.5×, HBM locked pricing) or NOW (fwd P/E ~21×, AI rerating thesis). Both offer superior risk-adjusted returns at current prices. BE in the portfolio makes sense only at entry, not at $277.
Thesis-breakers
▾
- 01 Manufacturing capacity misses 2 GW target by December 2026: if Q3 2026 unit shipments pace below run-rate for $3.4B revenue floor (implied units ~900–950 MW/quarter), the scale thesis breaks — production ramp failure at this valuation is catastrophic.
- 02 Oracle's initial 1.2 GW deployment defers by more than one quarter beyond 2026 commitment: any Oracle press release or earnings call referencing "revised deployment timeline" or "grid access secured" for Bloom-powered sites → thesis on concentrated hyperscaler demand is broken.
- 03 Non-GAAP gross margin fails to reach ≥32% in any quarter of H2 2026: below this level the FY2026 guidance of 34% is unreachable, signalling the product cost roadmap and manufacturing overhead absorption are not tracking — kills the path to 38–40% margin and FCF sustainability.
Key risks
▾
- 01 IRA/ITC policy risk: Bloom's 10-K explicitly cites the "One Big Beautiful Bill Act" as a material risk to Section 48E clean electricity ITC (30% credit, effective 2026). If credits are reduced or delayed, customer economics on PPA pricing deteriorate — potentially pausing the $2.65B AEP 20-year agreement and reducing Equinix economics. Magnitude: 20–30% derating of contracted pipeline.4
- 02 Oracle concentration risk: Oracle's 2.8 GW agreement (1.2 GW contracted) could represent 40–50% of FY2027 revenue if the growth trajectory holds. Any Oracle capex reprioritisation — OCI buildout pause, trade tariff impact on Oracle's global operations, or Stargate AI programme reallocation — creates a revenue gap Bloom cannot fill in 1–2 quarters. A single customer representing half of 2027 revenue at 135× P/E is a critical single point of failure.
- 03 Natural gas price exposure: SOFCs run on natural gas; if NG prices spike >30% from current ~$3.50/MMBtu, customer economics in long-term PPAs become unfavourable vs. solar+battery declining cost curves. This risk is partially mitigated by long-term supply agreements but not eliminated — and it's structurally absent in competitors like GEV (gas turbines with grid-scale economics) and NEE (renewables).
- 04 Convertible note dilution: $2.2B in 0% convertibles due 2030 at ~120% of stock price at issuance — if BE declines 40%+ from current levels, refinancing becomes materially more expensive or dilutive. The zero-coupon structure avoids near-term cash drain but creates a 2030 cliff that management must navigate.6
- 05 Beta risk at these multiples: Beta 3.2 means BE declines 3× as much as the broad market in a risk-off event. A 10% broad market correction implies ~32% BE drawdown — from $277, that's ~$188, coincidentally near the bottom of the entry zone. This is not an intrinsic risk to the thesis, but it means buy-and-hold from current levels involves tolerating extreme short-term volatility on a position not fundamentally supported at $277.
DCF scenarios
▾
Discount rate: 14% — recently GAAP-profitable (Q1 2026 first), high growth, beta 3.2, significant manufacturing execution risk, capital-intensive hardware (not SaaS). Method: EV/EBITDA terminal, 5-year EBITDA growth path from FY2026 guided midpoint ($725M adj. EBITDA).5
5Y EBITDA CAGR
50%
FY2031 EBITDA
$5.5B
Terminal EV/EBITDA
15×
Fair value / share
~$160
Bull requires: Oracle 2.8 GW fully deployed by end-2027; 2 new hyperscaler contracts of similar size by 2028; EBITDA margin expanding from ~20% (FY2026) to ~40% by FY2031 via manufacturing scale and service mix shift. Base: FY2026 EBITDA $725M → $1.09B → $1.63B → $2.45B → $3.67B → $5.50B. Terminal EV $82.5B; PV at 14% for 5 years = $42.9B; plus PV of interim EBITDA ~$3.5B; total EV ~$46.4B; less ~$120M net recourse debt; equity $46.3B ÷ 289M shares = ~$160/share.
5Y EBITDA CAGR
27%
FY2031 EBITDA
$2.4B
Terminal EV/EBITDA
12×
Fair value / share
~$57
Bear: 30% deceleration from Bull immediately — Oracle 2.8 GW partly defers, grid capacity improves faster in 2027, ITC credit uncertainty slows new signings, margin expansion stalls at 25% EBITDA. Base: FY2026 $725M → $921M → $1.17B → $1.49B → $1.89B → $2.40B. Terminal EV $28.8B; PV at 14% = $14.9B; plus PV of interim EBITDA ~$1.7B; total EV $16.6B; less debt; equity $16.5B ÷ 289M shares = ~$57/share.
Position: Current price $277 is above both the bull ($160) and bear ($57) DCF scenarios. The stock is fully priced beyond the fundamental bull case — consensus price targets ($236–$335) are anchoring on 2027 revenue multiples, not long-term FCF, which is why this disconnect persists.
Catalyst timeline
▾
-
2026-07Q2 2026 earnings (30 Jul): Management guided Q2 revenue "at least as good as Q1" ($751M+). If Q2 prints $750M+, it confirms the Oracle 1.2 GW deployment is on schedule and resolves execution risk — this is the most important near-term confirmation. A miss here (revenue below $650M) is a Dim 1 downgrade event.
-
2026-Q3Oracle deployment confirmation: Oracle is expected to confirm the first tranche of Bloom-powered data centres operational. Any Oracle earnings call or press release confirming AI workloads running on Bloom fuel cells validates the commercial model and could trigger analyst upgrades. Watch for Oracle's Q1 FY2027 earnings (around Sep 2026).
-
2026-H2New hyperscaler signings beyond Oracle: Microsoft and Equinix have existing agreements; AWS, Google, and Meta have not publicly committed. Any additional hyperscaler signing in H2 2026 would extend the demand runway and re-rate the FY2027 revenue line.
-
2026-Dec2 GW manufacturing capacity milestone: Bloom has targeted 2 GW annual capacity by December 2026. Confirmation or failure is the single most important operational data point of the year. Success unlocks FY2027 revenue guidance that could support the current multiple; failure forces a significant guidance cut.
Named analyst commentary
▾
"This leads me to my top pick for 2026 — Bloom Energy. Bloom Energy offers onsite power generation through solid oxide fuel cells that are behind the meter to reduce dependency on the grid. They can provide power in a few months, and they are significantly faster than the grid or nuclear. Bloom Energy is positioned as a mission-critical AI energy enabler, offering rapid, scalable on-site power solutions amid acute data center energy constraints."
— Beth Kindig / IO Fund (@bethkindig), Seeking Alpha, 2026-01-30 10
No coverage from tracked analysts: Serenity (@aleabitoreddit) and Dylan Patel / SemiAnalysis (@SemiAnalysis) — both focus on semiconductor/chips infrastructure, not AI Power. Searched; no relevant commentary found.
Sources
1 Robinhood / Investing.com — 52W range and current price as of 12–13 May 2026
2 Wells Fargo AI power demand projection, cited in Bloom Energy Seeking Alpha / IO Fund article, accessed May 2026
3 Bloom Energy IR / Bloomberg / enkiai.com analysis — Oracle 2.8 GW agreement (13 Apr 2026); AEP $2.65B agreement; Brookfield $5B financing; Equinix 100 MW+
4 SEC EDGAR — Bloom Energy Form 10-K / 8-K FY2025 risk factors; "One Big Beautiful Bill Act" ITC risk cited
5 Bloom Energy Q1 2026 Earnings Release (28 Apr 2026), Businesswire / SEC EDGAR 8-K; Benzinga full transcript
6 Bloom Energy press release — $2.2B convertible notes at 0% due 2030, priced Oct–Nov 2025; balance sheet as of Q1 2026 10-Q (recourse debt $2.60B, cash $2.52B)
7 SEC EDGAR Form 8-K — Bloom Energy appoints Simon Edwards as CFO, filed 26 Mar 2026
8 SEC EDGAR Form 4 filings — Soderberg: 42,881 shares sold 16–17 Mar at ~$150–161; 55,000 shares sold 14–15 Apr at $204–225; Joshi: 10,000 shares at $175.60 (Feb 2026); Bush: director sale (May 2026). All under 10b5-1 plans adopted 26 Nov 2025. Filing dates as sourced from stocktitan.net
9 GuruFocus — BE Forward PE Ratio 121.69× as of 10 May 2026; Seeking Alpha valuation metrics non-GAAP fwd PE 187.30×
10 Beth Kindig / IO Fund — "Bloom Energy: Solving The AI Data Center Power Bottleneck," Seeking Alpha, 2026-01-30
⚠ = figure not verifiable from a second open-access source or sourced from a single provider. Short interest data for BE was unavailable from Fintel free tier at time of research.
2 Wells Fargo AI power demand projection, cited in Bloom Energy Seeking Alpha / IO Fund article, accessed May 2026
3 Bloom Energy IR / Bloomberg / enkiai.com analysis — Oracle 2.8 GW agreement (13 Apr 2026); AEP $2.65B agreement; Brookfield $5B financing; Equinix 100 MW+
4 SEC EDGAR — Bloom Energy Form 10-K / 8-K FY2025 risk factors; "One Big Beautiful Bill Act" ITC risk cited
5 Bloom Energy Q1 2026 Earnings Release (28 Apr 2026), Businesswire / SEC EDGAR 8-K; Benzinga full transcript
6 Bloom Energy press release — $2.2B convertible notes at 0% due 2030, priced Oct–Nov 2025; balance sheet as of Q1 2026 10-Q (recourse debt $2.60B, cash $2.52B)
7 SEC EDGAR Form 8-K — Bloom Energy appoints Simon Edwards as CFO, filed 26 Mar 2026
8 SEC EDGAR Form 4 filings — Soderberg: 42,881 shares sold 16–17 Mar at ~$150–161; 55,000 shares sold 14–15 Apr at $204–225; Joshi: 10,000 shares at $175.60 (Feb 2026); Bush: director sale (May 2026). All under 10b5-1 plans adopted 26 Nov 2025. Filing dates as sourced from stocktitan.net
9 GuruFocus — BE Forward PE Ratio 121.69× as of 10 May 2026; Seeking Alpha valuation metrics non-GAAP fwd PE 187.30×
10 Beth Kindig / IO Fund — "Bloom Energy: Solving The AI Data Center Power Bottleneck," Seeking Alpha, 2026-01-30
⚠ = figure not verifiable from a second open-access source or sourced from a single provider. Short interest data for BE was unavailable from Fintel free tier at time of research.