NVDA Q2 Guide of $91B Beats Consensus, but China Exclusion Reframes the Bull Case

Nvidia's first-quarter results delivered the headline earnings beat that Wall Street anticipated, but the stock's muted after-hours reaction exposed a deeper unease: the narrative around AI infrastructure spending is shifting from pure euphoria to hard questions about unit economics and competitive dynamics. Revenue of $81.6 billion, up 85% year-over-year, and data center sales that nearly doubled underscored continued demand for Blackwell GPUs. Yet the market's hesitation signals that consensus has already priced in the company's near-term dominance.

The real debate centers on whether hyperscalers can profitably deploy the scale of chips Nvidia is prepared to ship. Amazon disclosed plans to add more than 1 million Blackwell and Rubin GPUs this year, translating to roughly $30-40 billion in capex just for that upgrade. Google and Baidu are similarly aggressive. Nvidia's own guidanceCompany-issued forecasts of future financial performance. of $91 billion in Q2 revenue, a 94% year-over-year growth rate, assumes this spending continues unabated. But margins are tightening as cloud operators build alternative silicon and negotiate harder on pricing. H100 rental prices are up roughly 20% even though that GPU generation is now three years old, suggesting supply constraints are easing and competitive pressure mounting.

The geopolitical backdrop compounds the uncertainty. Nvidia's Q2 guidanceCompany-issued forecasts of future financial performance. excludes any contribution from China's data center segment, a conscious acknowledgment of regulatory risk. This carveout reveals how much of the company's growth narrative depends on an increasingly fragile export regime. Meanwhile, institutional positioning looked uniformly bullish heading into earnings, sell-side consensus was bullish, retail flow was heavy long, and options implied volatilityThe market's forecast of future volatility, extracted from option prices. skewed toward upside surprises. A clean beat with no guide raise may be the worst outcome for asymmetric risk.

The broader implication is that AI infrastructure spending is decoupling from AI adoption. Worker-level productivity gains from large language models are visible, but enterprise-wide deployment remains experimental. If hyperscalers hit capex fatigue or if their custom silicon roadmaps accelerate, Nvidia's growth rate will compress faster than most models assume. The stock's immediate slide suggests the market is finally pricing that tail risk.