Nvidia Says It’s Getting Orders From China

Nvidia says it is now receiving orders from China and is ramping sales of its H200 AI chips, indicating renewed demand and potential near-term revenue upside for its data-center business. CEO Jensen Huang calling OpenClaw "the next ChatGPT" lifted Chinese AI equities on the comment, signaling positive sentiment across the AI sector. Separately, Kalshi CEO Tarek Mansour gave an interview after Arizona filed criminal charges against the prediction-market platform, creating a firm-specific regulatory/legal risk. Expect meaningful moves at the company and sector level (Nvidia, Chinese AI names, Kalshi) rather than a market-wide shock.

Analysis

Primary second-order beneficiaries are the specialized data‑center supply chain nodes: HBM memory vendors and high‑speed interconnect/network ASIC suppliers will see revenue pull‑through even if unit GPU ASPs moderate, because newer models drive higher HBM capacity and more ports per board. Systems OEMs that can quickly assemble and certify heterogeneous racks (SMCI, HPE equivalents) pick up the near‑term services and install revenue; conversely, GPU challengers (AMD/Intel and nascent Chinese silicon) face compressed ramp windows and must undercut on price or accept narrower design wins.

Key catalysts and tail risks bifurcate by horizon. Over the next 1–3 quarters, order flows and inventory digestion across hyperscalers will drive volatility; watch shipment/booking cadence and channel inventory metrics as early signals. Over 12–36 months the structural risk is geopolitical: meaningful new export controls or rapid Chinese silicon substitution could truncate TAM and reprice growth multiples; conversely, sustained enterprise AI adoption would expand FCF materially and justify premium multiple expansion.

Consensus is underestimating asymmetry between near‑term pricing power and long‑term localization risk. The market gives large credit for perpetual scale economies but underweights a scenario where China substitutes at 20–40% of local demand within 24–36 months, capping global pricing power. That creates a sweet spot for directional trades that capture near‑term growth with defined downside protection against a geopolitical shock.

Key Decisions for Investors

• Directional, defined‑risk: Buy a 6‑9 month NVDA call spread (long ATM, short ~+30% OTM) sized to 1–2% NAV. Rationale: captures near‑term upside from continued AI demand while capping premium spend; target >2.5x upside if shares rally 30%+, max loss = premium paid; scale out on 30–50% realized move.
• Pair trade to de‑risk market beta: Long NVDA equity / Short AMD equal notional (3–9 month horizon). Rationale: isolates product‑cycle share gains. Target spread widening of 15–25% in NVDA’s favor; stop if relative moves exceed 30% against position.
• Supply‑chain exposure: Go long MRVL or AVGO (6–12 months) on weakness into earnings to play network/HPC pull‑through. Rationale: networking revenue lags GPU ramps by 1–2 quarters and carries higher margin; position size 1–3% NAV with 20–30% upside target.
• Tail hedge: Buy 9–12 month OTM NVDA puts (10–20% OTM) sized to cover 50% of upside exposure above. Rationale: protects against an abrupt export‑control or China substitution shock; acceptable cost as insurance vs concentrated directional exposure.