The Tech Download: What you might have missed in Nvidia’s earnings — a $200 billion opportunity and edge computing

Nvidia reported revenue up 85% to $82 billion, announced an $80 billion share buyback, and raised its dividend, underscoring another blockbuster quarter. Management flagged China as a major headwind, saying Nvidia has "largely conceded" the AI chip market there to Huawei amid U.S. export controls and Beijing's push for domestic alternatives. The company also reframed reporting into data center and edge computing and said its new Vera CPU opens a $200 billion market, with $20 billion in expected CPU revenue this year.

Analysis

The key signal is not the headline quarter; it is the product mix shift. Nvidia is effectively trying to re-rate itself from a single-bet data-center GPU supplier into a broader AI compute stack provider, which should widen the investor base and soften cyclicality over time. The market’s muted reaction suggests skepticism that the incremental CPU and edge narratives can offset the concentration risk embedded in hyperscaler capex, especially if AI spending growth normalizes after the current buildout phase.

China is the hidden margin variable. If Nvidia has truly ceded meaningful share to domestic alternatives, the near-term revenue hit may be manageable, but the bigger second-order effect is mix degradation: lower access to the highest-growth geography plus a greater reliance on a small set of U.S. cloud buyers increases bargaining pressure on price and roadmap. That also creates a relative beneficiary set in China-centric supply chains and domestic silicon names, while reducing the likelihood that easing export rules alone can re-accelerate Nvidia’s China franchise.

The buyback matters less as a signal of undervaluation and more as a capital-allocation floor that can dampen drawdowns, but it will not change the fact that the stock is trading against very high expectations on gross margin durability and inference monetization. Over the next 3-6 months, the main catalyst is not another beat; it is whether the market starts to believe CPU and edge can meaningfully extend the growth curve beyond the current GPU cycle. If not, any rotation out of mega-cap AI spend or a negative read-through from hyperscaler capex could compress the multiple even while fundamentals remain excellent.

The contrarian view is that investors are underestimating how much of the AI stack Nvidia can monetize outside GPUs if inference becomes the dominant workload. That argues for patience on the long side, but only if one is willing to tolerate valuation compression in the interim; this is a good business with a crowded ownership base, not a low-risk trade at current sentiment.

Key Decisions for Investors

• Maintain a core long NVDA, but hedge with short-dated puts into the next 4-8 weeks of hyperscaler capex commentary; the setup favors earnings resilience but multiple compression if AI spend growth decelerates.
• Pair trade: long NVDA / short AMD over 3-6 months if you believe Nvidia’s platform expansion into CPU and edge is more credible; AMD has more narrative leverage but less demonstrated pricing power.
• Add selectively on NVDA weakness only if it trades down on no change to forward estimates; use 5-10% pullbacks as entry points rather than chasing post-earnings strength.
• Watch China-exposed semiconductor names for relative value longs over the next 1-2 quarters; if domestic substitution persists, the second-order beneficiary is local AI silicon, not Nvidia.
• For event-driven risk control, reduce gross exposure if the next two hyperscaler earnings reports show capex discipline; that is the clearest catalyst for a 10-15% de-rating in the AI complex.