EHang Holdings

Phase A — Understand the business

Lens 1 · Company Overview

EHang is a Guangzhou-based advanced air mobility (AAM) company that designs, builds, sells and operates pilotless (fully autonomous, no onboard pilot) electric vertical-takeoff-and-landing (eVTOL) aircraft. Its flagship is the EH216-S, a two-seat passenger autonomous aerial vehicle. The business model is hardware sales (aircraft units) plus a growing services layer (MRO, operation training, flight demonstration, aerial media performances, vertiport planning).

Revenue mix FY2025: air mobility solutions RMB395.2M (94.5%) and others RMB22.8M (5.5%). The core is aircraft deliveries: 169 EH216-series + 6 VT35 units in 2025 (vs 216 EH216 in 2024). Cumulatively 628 eVTOL units delivered since 2018 (593 EH216-S, 17 EH216-F, 12 EH216-L, 6 VT35), "mainly operated on a limited trial basis in tourism locations in China, for testing, training, demonstration and trial operation purposes" — the company's own words.

Customers: tourism/scenic-area operators in China (the dominant use case today), plus international buyers via framework/sales agreements (e.g. UAE's Wings Logistics Hub, which signed for up to 100 units ). Suppliers: outsourced contract manufacturers + single-source components for some parts; battery technology is explicitly flagged as "beyond our control". Competitors: Joby, Archer (US, piloted, FAA track); AutoFlight, XPeng AeroHT (China); plus the whole question of whether ground transport / helicopters are the real substitute.

Contract structure / payment terms (key): for air-mobility and aerial-media, EHang "typically requires a portion of payments upfront and the remaining amounts are contractually due ranging from three to six months". This is NOT take-or-pay and NOT recurring — it's discrete unit sales with deferred receivables, which is why accounts receivable ballooned (see Lens 10). The headline "pre-order backlog" of 1,300+ units is built on framework agreements that "do not obligate our counterparties to purchase our products at all" — a critical, self-disclosed caveat.

Lens 2 · Supply Chain

Map: upstream components → EHang (design + final assembly, partly outsourced) → operator/end-customer.

• Upstream inputs: carbon-fibre composite airframe materials; electric motors / propulsion; batteries (the binding constraint — energy density caps the 30-km / ~25-min range of the EH216-S); avionics, flight-control computers, sensors, communication/data links; ground control stations. The 20-F states "some of the components used in our products are currently selected to be purchased from a single source to improve cost-efficiency" and that performance specs "dependent on battery technology" are "constrained by the pace of general technological advancement and the capabilities of our suppliers, which are largely beyond our control". Specific component suppliers are not named in the 20-F — n/a — not disclosed at supplier-name granularity. This is a real gap: unlike a chip-supply-chain dive, EHang does not disclose its bill-of-materials vendors.
• Manufacturing: in-house facilities at Yunfu (manufacturing) and Guangzhou (EHang Future City HQ); capacity expansion underway (FY2025 capex RMB160.4M, up from RMB40.1M, "for the establishment and installation of equipment at our manufacturing facility in Yunfu" + new Guangzhou office building). Certain manufacturing activities are outsourced to third-party contract manufacturers.
• Downstream: scenic-area / tourism operators (China); JV operating companies (Guangdong EHang General Aviation; Hefei Heyi Aviation — the JV that holds an Air Operator Certificate); international distributors/partners (Wings Logistics Hub UAE; Spain Lleida-Alguaire UAM Center; Mexico; Thailand; Brazil).
• Chokepoints: (1) battery energy density = hard range ceiling; (2) single-source components; (3) CAAC certification is itself a chokepoint that EHang has cleared and rivals have not (a moat, see Lens 3); (4) contract manufacturing quality control for a safety-critical human-carrying product.

Lens 3 · Competitive Advantages (moats)

The one real moat is regulatory first-mover status. EHang holds the world's first full triple-crown for a human-carrying pilotless eVTOL from the CAAC: Type Certificate (Oct 2023), Production Certificate (Mar 2024), and Air Operator Certificate (Mar 2025). No other company globally has all three for an autonomous passenger eVTOL. By contrast:

• Joby: FAA Stage 4 of 5, entered Type Inspection Authorization (TIA) March 2026, TC estimated late 2026 — and it's piloted, not autonomous.
• Archer: FAA Stage 3, targeting cert before LA 2028 Olympics.

So on the certification axis EHang is genuinely 2–3 years ahead. This is durable while it lasts: certification is slow, expensive, and a true barrier to entry.

But the moat is narrower than the bull case implies:

1. It's CAAC-only. A CAAC TC does not transfer to FAA or EASA. EHang's certification advantage is a China advantage, not a global one. In the US/EU, Joby/Archer are ahead.
2. The autonomy advantage cuts both ways. Pilotless = lower opex and no pilot-supply constraint (a real edge), but also a higher public-trust and regulatory bar outside China. Western regulators are far more conservative on no-pilot passenger flight.
3. Brand/perceived value: strong within China's "low-altitude economy" policy narrative; thin globally.
4. Bargaining power: weak over customers (24% revenue concentration, framework orders non-binding) and weak over suppliers (single-source, battery-constrained). EHang needs its customers more than they need EHang — the opposite of a moat.
5. AutoFlight reportedly now "holds all three certifications required for commercial operations" for its CarryAll cargo eVTOL (crossed the Qiongzhou Strait Dec 31, 2025, 400kg/200km). EHang is no longer the only certified Chinese eVTOL — the moat is already eroding at the edges.

Verdict on moat: real but time-boxed and geographically boxed. It is a 2–3 year head-start in one country, not a structural monopoly.

Lens 4 · Segments

EHang reports two revenue lines (no formal product/geographic segment P&L in the 20-F — geography revenue is overwhelmingly China-domestic and not broken out into a segment table):

Trend & cause — decelerating, then contracting: revenue exploded 3.9x in 2024 (52 → 216 units as certification de-risked sales), then fell 8.4% in 2025 as unit volume dropped to 169 EH216 + 6 VT35. The core air-mobility line shrank (-10.9% YoY). The only growth was "others" (+77%, off a tiny base, driven by smart-city + aerial media). Most alarming forward signal: in Q1 2026, air-mobility deliveries collapsed to just 4 EH216 units (vs 11 in Q1'25), and aerial media was ~40% of the quarter's RMB25.7M revenue. The product that is supposed to be the growth story is, on a trailing-quarter basis, the minority of revenue and shrinking. Management's FY2026 guidance of ~RMB600M (≈+43% YoY) is therefore extraordinarily back-half-loaded against a RMB25.7M Q1 — credibility risk is high (see Lens 11/13).

---

Phase B — Measure performance

Lens 5 · Earnings Result (latest print)

Two prints matter: the audited FY2025 20-F, and the Q1 2026 6-K.

FY2025 (audited):

• Revenue RMB418.0M (US$59.8M), −8.4% YoY
• Cost of revenue RMB160.8M → Gross profit RMB257.2M, gross margin 61.5% (flat vs 61.4% in 2024 — margin is the bright spot)
• Operating expenses RMB586.2M (140.3% of revenue!) — G&A alone RMB269.6M EXCEEDS total revenue; R&D RMB194.6M; S&M RMB122.0M
• Share-based compensation RMB246.2M (US$35.2M) = 59% of revenue — the dominant cost driver
• Operating loss RMB316.7M (US$45.3M), worse than 2024's RMB254.1M
• Net loss RMB276.4M (US$39.5M), vs RMB230.0M in 2024 (loss widened 20.2%)
• Interest & investment income RMB58.6M (a real offset — they earn yield on the cash pile)
• One-off: US$1.985M securities-litigation settlement (Pujo, see Lens 10) booked in non-operating expense

Q1 2026 (unaudited):

• Revenue RMB25.7M (4 EH216 units), gross margin 62.5%
• Operating loss RMB127.9M, net loss RMB126.4M (loss widened YoY)
• $30M share buyback announced over 12 months
• FY2026 guidance maintained at ~RMB600M

Balance-sheet flags: cash & equivalents fell from RMB610.9M to RMB256.4M, but short-term investments rose from RMB513.7M to RMB843.2M — they shifted cash into yield instruments, not a liquidity loss. Total liquidity (cash + ST investments + restricted) = RMB1,129.3M (US$161.5M) at YE2025. BUT operating cash flow swung from +RMB158.0M (2024) to −RMB179.5M (US$25.7M) (2025) — the company burned cash from operations again after one positive year. Accounts receivable jumped to RMB106.5M and inventory to RMB101.6M (both growing faster than the shrinking revenue — see Lens 10). Market reaction: the stock is down ~61% over the trailing year to ~$6.13 (mkt cap ~$465M), near its 52-week low of ~$6.50. The tape says the market does not believe the RMB600M guidance.

Lens 6 · Earnings Calls (sentiment trend)

No transcripts on the research shelf (transcripts/ empty) — this lens is ``-grounded.

• FY2024 call (early 2025): euphoric — "world's first full suite of certifications," first-ever positive adjusted net income in Q4'24, record EH216-S sales. Management framed 2025 as the commercial-scaling year.
• Through 2025: tone shifted to "building a replicable and promotable business model" and "expanding the partnership network" — softer, more process-oriented language as actual deliveries decelerated.
• Q1 2026 call: defensive — emphasis on VT35 certification progress (still only in "Certification Basis definition phase"), aerial-media contribution (~40% of revenue), international expansion (Thailand, Mexico), and a buyback to signal confidence amid a 52-week low.

Things they stopped saying: the aggressive unit-backlog conversion narrative has quietly receded; the framing has moved from "we have 1,300 orders" to "we are building a business model." That linguistic shift — from backlog to model-building — is itself a tell that the Hindenburg "hollow order book" critique landed.

Lens 7 · Comps

EHang's only true public pure-play peers are Western eVTOL developers (different geography, both pre-revenue on aircraft). The comparison is the single most important table in this dossier because it frames the entire bull case.

P/E, dividend yield, 5-yr avg ROE: n/a — all three companies are loss-making with negative equity returns and pay no dividend. A P/E does not exist for any of them.

The provenance-honest read: EHang trades at ~1/16th of Joby's market cap while generating more aircraft revenue than Joby and Archer combined (Joby's $53.4M is largely Blade helicopter ops, not eVTOL) — at 61.5% gross margin, which is remarkable for hardware. On EV/Sales, EHang at ~5x vs Joby at ~140x is a ~25x relative-valuation gap. If you believe EHang's revenue is real and CAAC certification has value, EHang is the cheapest certified eVTOL exposure on the market by an order of magnitude. If you believe Hindenburg, the revenue quality is the question and the discount is deserved. The comps table does not resolve the thesis — it sharpens it to a single binary: is the revenue real and is the China-only cert valuable?

Lens 8 · Stock-Price Catalysts (>5% moves, last ~5 years)

Pattern is overwhelmingly binary-event driven, not earnings-driven:

• Nov 7, 2023 — Hindenburg short report: −16%+ in a session ("Hollow order book and fake sales") — the single largest single-day catalyst.
• Oct 2023 — CAAC Type Certificate: sharp rally (world-first milestone).
• Apr 2024 — Production Certificate: rally.
• Mar 2025 — Air Operator Certificate granted to operators: rally (full triple-crown).
• 2024–25 — ATM equity raises / dilution events: persistent overhang (US$76.2M raised via ATM in 2024, US$23.8M in 2025) — dilution is a recurring negative catalyst.
• Feb 2026 — 52-week low ~$6.50: "commercialization push" not converting; UBS downgrade Buy→Neutral with 50–70% estimate cuts.
• Q1 2026 — weak print + $30M buyback: mixed.
• Net: down ~61% over the trailing year.

What the market actually reacts to: (1) CAAC regulatory milestones (up), (2) short-seller / delivery-disappointment / dilution (down), (3) China low-altitude-economy policy headlines (up). It does NOT trade on quarterly EPS in the conventional sense — it trades on the certification-and-commercialization narrative and credibility. This is a story stock priced on a binary commercialization outcome.

---

Phase C — Judge people & books

Lens 9 · Management

• Huazhi Hu — Founder, CEO & Chairman since inception. Founded predecessor Beijing Yihang Chuangshi (large-scale command-and-control systems) in 2005; ex-CTO of Beijing 999 Emergency Rescue Center; studied computer science at Tsinghua (1992–97). Founder archetype, technically credible, deeply entrenched. Controls 77.9% of voting power via 39.0M Class B super-voting shares (10 votes each) through his BVI vehicle Genesis Rising Limited, despite only 26.6% economic ownership. Skin in the game: meaningful (26.6% economic). Governance control: absolute — minority ADR holders cannot meaningfully challenge the board. Dual-class + China VIE = ADR holders are about as far from control as it is possible to be.
• Conor Chia-hung Yang — CFO since Sept 2023, director since 2019. Ex-Goldman Sachs, Lehman, Morgan Stanley (Asia) banker; serial China-ADR CFO (Tuniu/TOUR, Dangdang, AirMedia). Sits on 4 other Nasdaq/HK boards (NovaBridge, iQIYI, Tongcheng, UP Fintech) — a professional, capital-markets-savvy CFO, but a portfolio CFO whose attention is split. He was hired ~5 weeks before the Hindenburg report — i.e. brought in to professionalize the capital-markets story.
• Board: notable presence of Dongming Wu, CEO of DHL Express China (a logistics tie-in that supports the cargo-eVTOL thesis), and VC Haoxiang Hou (Forbes China Top-100 VC).
• Capital-allocation history: mixed-to-poor. The company has funded itself with serial dilution (US$95.6M private placements 2021–24 + US$100M ATM 2024–25) and increasingly with debt (FY2025 financing: RMB249M short-term + RMB75M long-term bank loans). It just launched a $30M buyback while still burning operating cash — arguably a defensive price-support move that contradicts the "we need to raise funds" going-concern language. ROE/ROIC: deeply negative (accumulated deficit RMB2,262.4M / US$323.5M). On their watch, no value has been returned; the bet is entirely on future commercialization.
• Red flags: (1) the Hindenburg report directly alleged management misled on the order book — the most serious management-integrity question; (2) SBC at 59% of revenue is shareholder-unfriendly; (3) dual-class entrenchment; (4) a settled (not litigated-to-victory) securities class action.

Lens 10 · Forensic Red Flags

Acting as a forensic equity analyst. This is the lens that matters most for EHang.

Income statement:

• Share-based compensation = RMB246.2M (US$35.2M), 59% of revenue. Non-GAAP "adjusted" profitability (which management touted for Q4'24) is manufactured almost entirely by adding SBC back. Treat any "adjusted net income" claim with extreme skepticism — on a GAAP basis the company lost RMB276.4M.
• Operating expenses (140% of revenue) dwarf the revenue base; G&A alone exceeds total revenue. This is not an operating company at scale; it is an R&D/commercialization project with some sales.

Balance sheet — the classic concentration + receivables flag (which the 20-F itself lists as a named risk):

• Accounts receivable rose to RMB106.5M (2025) from RMB58.2M (2024) — +83% — while revenue FELL 8.4%. Receivables growing as revenue shrinks is a canonical revenue-quality warning. The 20-F lists "we have experienced a significant increase in accounts receivable" as a standalone risk factor.
• Customer concentration: largest customer = 24% of revenue (RMB100M / US$14.3M) in 2025 (26% in 2024, 24% in 2023); top-3 ≈ RMB235M ≈ 56% of revenue. AR from top customers = 51% of total AR balance. If one tourism-operator customer stops paying or buying, the P&L breaks.
• Inventory rose to RMB101.6M from RMB75.7M (+34%) against falling sales — inventory building ahead of guidance that Q1 undercut.
• Contract liabilities (customer advances) FELL to RMB60.8M from RMB62.6M — deferred revenue is not building, which would normally precede a demand acceleration. A soft leading indicator against the RMB600M guide.

Cash flow vs earnings: operating cash flow −RMB179.5M while net loss was −RMB276.4M — the gap is largely SBC (non-cash) offset by working-capital drag (rising AR/inventory consume cash). The 2024 operating inflow (+RMB158M) was an anomaly driven by customer advances on the certification-year order surge; 2025 reverted to cash burn. Do not extrapolate 2024.

Going concern: the 20-F includes explicit going-concern framing — financials "prepared assuming we will continue as a going concern… Our ability to continue as a going concern is largely dependent on successful execution of our business plan to scale up commercial sales of EH216-S and our ability to raise additional funds when needed". They assert liquidity is "sufficient… for at least the next 12 months" — standard, but the dependence on raising more is stated plainly. With ~US$161.5M liquidity and ~US$26M annual operating burn (pre-capex; capex was another US$22.9M), runway is real but not infinite, and it shrinks if the RMB600M guide misses and capex stays elevated.

VIE / China-ADR structure: EHang is a Cayman Islands holding company that consolidates a PRC VIE (the operating entity) via contractual arrangements, with the WFOE in between. ADR holders own the Cayman shell, not the operating business. Risks: (1) PRC could rule the VIE contracts non-compliant; (2) cash is trapped — only 24.2% of liquidity is held in the PRC and dividends up-streaming face SAFE/forex controls; (3) no dividends ever paid and none planned; (4) HFCAA/PCAOB delisting risk (China-ADR audit-access regime). These are structural, not idiosyncratic, but they materially cap the multiple a Western investor should pay.

Regulatory findings (required sub-section):

• SEC EDGAR EFTS (LR + AAER): No Litigation Releases and no AAERs naming EHang in the search period 2021-06-30 → 2026-06-30.
• 10-K/20-F Item 8 Legal Proceedings (self-disclosed): Pujo v. EHang Holdings Limited (No. 2:23-cv-10165, C.D. Cal.) — securities class action filed Dec 4, 2023, asserting Section 10(b)/20(a) claims, "based in large part on a November 7, 2023 report issued by short seller Hindenburg Research," alleging false/misleading statements about business, operations and prospects. The parties settled for US$1,985 thousand; settlement executed Aug 11, 2025; final court approval Jan 12, 2026; settlement fully paid as of Dec 31, 2025. Settlement is not an admission, but a company that paid ~$2M to make a Hindenburg-driven suit disappear is not a clean bill of health.
• Non-SEC enforcement (web search): no material FTC/DOJ/FDA/CFPB enforcement actions, consent decrees, or fines against EHang surfaced. The Hindenburg Research report (Nov 7, 2023) — title "EHang: Hollow order book and fake sales make this China-based eVTOL company last in line for takeoff" — alleged >92% of the 1,300+ unit pre-order book was "dead" or "abandoned" and that revenue had "all the hallmarks of fake revenue," including a former employee calling a United-Therapeutics-linked preorder "dead." EHang denied all allegations as "untrue statements and misinterpretations". This is a short-seller allegation, not a regulatory finding, and no regulator has charged EHang — but it is the central forensic question and the genesis of the (now-settled) class action.
• Summary: No SEC enforcement; one settled securities class action rooted in a short-seller report; the order-book-quality allegation remains unadjudicated and is partially corroborated by the 20-F's own admission that framework orders "do not obligate counterparties to purchase at all."

---

Phase D — Project & stress-test

Lens 11 · Forward Projection (FY2026 / FY2027 / FY2028 EPS)

EHang is loss-making, so the meaningful projection is revenue → operating loss → net loss per ADS, built bottom-up from FY2025 actuals + the RMB600M FY2026 guide. Shares ≈ 150.2M ordinary = ~75.1M ADS (each ADS = 2 Class A shares).

Base inputs (labeled):

• FY2025 actual: revenue RMB418.0M, gross margin 61.5%, opex RMB586.2M, net loss RMB276.4M.
• FY2026 revenue: management guides ~RMB600M; but Q1'26 was only RMB25.7M, requiring ~RMB574M across Q2–Q4 — a ~7.5x H2 ramp. I haircut the guide.

. FY27 base ≈ −US$0.42/ADS; FY28 base ≈ −US$0.34/ADS.

Per-ADS EPS, base case: FY2026 ≈ −$0.52 · FY2027 ≈ −$0.42 · FY2028 ≈ −$0.34 (all ``, GAAP, loss-narrowing but negative through the projection window). A GAAP profit is not a FY2026–2028 base-case event; the bull case reaches near operating breakeven only in FY2028 and only if the RMB600M guide is met and held.

Brier forecast NOT logged — per --watchlist rules (no forecast.ts create in the breadth loop; only log when genuinely committing to a base case). The natural scoreable forecast to log on a future interactive pass: "EH FY2026 revenue ≥ RMB550M" — p ≈ 0.30 (i.e. I expect the guide to miss).

Lens 12 · Bull vs Bear

Bull case. EHang is the only certified, revenue-generating autonomous passenger eVTOL on Earth, trading at ~$465M — a fraction of pre-revenue Joby ($7.6B+) and Archer ($5B+). It owns a 2–3 year CAAC certification lead, 61.5% gross margins (hardware margins most automakers would envy), and sits at the centre of China's trillion-yuan "low-altitude economy" — a named priority in the 15th Five-Year Plan (2026–30), with Shenzhen alone funding 1,200+ vertiports and ¥12B. As commercial sightseeing operations scale across Guangzhou/Hefei and exports ramp (UAE, Brazil, Spain), unit volume re-accelerates, operating leverage kicks in, and the stock re-rates from "fraud-discount micro-cap" to "the Tesla of the sky." The $30M buyback signals the founder thinks shares are mispriced. Potential surprise: a single large binding government/tourism fleet order converts the framework backlog into real revenue and validates the whole story overnight.

Bear case (2–3 permanent-impairment risks).

1. The demand is a tourism novelty, not transportation, and it's stalling. The 20-F concedes aircraft are "mainly operated on a limited trial basis in tourism locations." FY2025 revenue fell; Q1'26 deliveries collapsed to 4 units and aerial-media (drones, not air taxis) became 40% of revenue. If pilotless air-taxis are a scenic-area gimmick rather than a mobility platform, the TAM is a rounding error and the multiple is fiction.
2. Revenue quality / order-book credibility. Hindenburg alleged the backlog is "smoke" and revenue is "fake"; the 20-F admits framework orders don't obligate anyone to buy; receivables are up 83% as revenue falls; customer concentration is 24–56%. A revenue restatement or a major customer default would be terminal.
3. China-VIE + delisting + trapped cash. ADR holders own a Cayman shell over a PRC VIE; cash is forex-trapped; HFCAA/PCAOB and geopolitical delisting risk is a permanent tail.

Pre-mortem (18 months out, thesis broke): FY2026 revenue came in ~RMB420M (missed the RMB600M guide badly); Q1's 4-unit print proved to be the trend, not a seasonal dip; AutoFlight and XPeng AeroHT won the cargo and intra-city share in China; a second short report or an SEC comment letter reopened the order-book question; the company did another dilutive ATM raise at a lower price; and the stock halved again to ~$3 on "story broken, still burning."

Are multiples too high? On EV/Sales (~5x) EHang is cheap vs eVTOL peers but not cheap vs its own fundamentals — a shrinking-revenue, cash-burning, 140%-opex micro-cap with a China discount arguably deserves a low-single-digit sales multiple. The Joby/Archer comps are themselves in a bubble; being cheaper than a bubble is not "cheap."

Contrarian view (what the market refuses to see): Both bulls and bears are fighting the 2023 war (Hindenburg fraud vs. certification triumph). The thing the market is missing is mundane: EHang's actual 2025–26 numbers show the air-taxi business shrinking while a drone-light-show business grows — i.e. EHang may quietly be becoming an aerial-media and smart-city drone company that also happens to hold an eVTOL type certificate, not the passenger-mobility platform its valuation narrative requires. If so, it's neither a fraud nor a 100-bagger — it's a niche Chinese commercial-drone company worth roughly… its current ~$465M.

Lens 13 · Devil's Advocate (short-seller)

Dismantling the bull case.

• What structurally breaks the model: the core product (passenger eVTOL) is not selling — 169 units in 2025 down from 216, and 4 in Q1'26. A "growth company" whose flagship volume is falling has no model to break; it's already breaking. Revenue is propped by low-margin aerial-media drones and lumpy smart-city deals.
• Revenue concentration: 24% in one customer, 56% in three, 51% of receivables in the top customers — all tourism operators in China buying for scenic rides. This is not a diversified, recurring revenue base; it's a handful of related-looking buyers in a single use case Hindenburg alleged were "newly formed customer entities with no discernible operations."
• Why the moat is weaker than bulls think: the CAAC cert is non-transferable (worthless for FAA/EASA markets) and no longer unique (AutoFlight reportedly has all three certs for cargo). The "first-mover" advantage is a first-mover advantage in certification paperwork in one country, not in demand.
• Most dangerous underestimated competitor: XPeng AeroHT — backed by XPeng's auto balance sheet, manufacturing scale, and consumer brand; targeting modular flying cars for the same China low-altitude market with vastly deeper pockets than EHang's $465M shell. Also AutoFlight (already cross-strait certified, cargo). EHang could be out-capitalized in its home market.
• Worst capital-allocation / accounting flags: SBC at 59% of revenue funds insiders while diluting ADR holders; serial ATM dilution at falling prices; a $30M buyback launched while burning operating cash and warning it may need to raise more (incoherent); adjusted-profit framing that exists only by adding back SBC.
• Assumptions that must hold for today's ~$6 price: (1) the RMB600M FY2026 guide is roughly met; (2) CAAC commercial operations actually monetize beyond scenic rides; (3) no second short report / no SEC inquiry / no delisting escalation; (4) the founder's 78% control is used benignly.
• If growth disappoints 20–30%: at RMB420M FY2026 (a 30% guide miss), there is no operating-leverage story, runway shortens, a dilutive raise becomes likely within 12–18 months, and the stock re-rates toward cash value (~$2–3).
• Single scenario that permanently impairs: a revenue-recognition restatement or SEC enforcement validating the Hindenburg "fake sales" thesis — plausibility low-to-moderate (no regulator has acted in 2.5 years, and the class action settled cheaply), but the tail is fat because of the VIE structure and the company's own non-binding-order disclosures.

Lens 14 · Management Questions (ordered by information value)

1. Q1 2026 delivered 4 EH216 units vs 11 a year earlier, yet you maintain ~RMB600M FY2026 guidance. What specific binding (not framework) orders and delivery schedule bridge a RMB25.7M Q1 to RMB600M for the year — customer by customer?
2. Of the 1,300+ unit pre-order/backlog historically cited, how many are firm, deposit-backed purchase orders vs non-binding framework MOUs today — and what is the dollar value of deposits actually received?
3. Your largest customer is 24% of revenue and the top three ~56%; accounts receivable rose 83% while revenue fell 8%. Who are these customers, what is their independent operating business, and what is the aging/collectibility of that receivable?
4. Air-mobility revenue is shrinking while aerial-media is now ~40% of quarterly revenue. Is EHang becoming a commercial-drone/aerial-media company? What is the realistic 3-year revenue split between passenger eVTOL and everything else?
5. You launched a $30M buyback while reporting going-concern dependence on raising additional funds. Reconcile buying back stock with potentially needing to issue stock.
6. CAAC certification is non-transferable to FAA/EASA. What is the concrete, funded timeline and capital plan to certify in any market outside China, and what revenue do you underwrite from exports before then?
7. AutoFlight reportedly holds all three certifications for cargo eVTOL and XPeng AeroHT is entering with auto-scale capital. How is your China moat durable against better-capitalized domestic rivals?
8. SBC was 59% of revenue (RMB246M). What is the multi-year plan to bring SBC to a normal percentage of revenue, and how do you justify the current level to diluted ADR holders?
9. The VT35 is only in the "Certification Basis definition phase." What is the realistic year of VT35 type certification and first commercial revenue — and what has been spent to date?
10. Operating cash flow swung from +RMB158M (2024) to −RMB179M (2025). What is your quarterly cash-burn run-rate for 2026 including capex, and at what date does liquidity fall below 12 months at the current burn?
11. What unit economics does a scenic-area operator actually achieve per EH216-S (utilization, ticket price, payback period)? Show one real operator's P&L.
12. How much of the RMB1,129M liquidity is upstreamable to the Cayman holdco given SAFE/forex controls, and what is your contingency if PRC capital controls tighten?
13. What is your exposure to HFCAA/PCAOB delisting and what concrete steps protect ADR holders if US-China audit access deteriorates?
14. Battery energy density caps the EH216-S at ~30 km / ~25 min. What is the roadmap to a range that enables genuine point-to-point urban transport (vs scenic loops), and who supplies that battery?
15. Founder voting control is 77.9%. What independent-board and minority-protection mechanisms exist to prevent related-party value leakage, given the prior short-seller allegations?

---