Nvidia's Valuation Just Hit Lows Not Seen Since 2019, Even as Its Profits Doubled

Nvidia reported fiscal Q2 revenue of $96.2B (+106% YoY, +18% QoQ) and adjusted EPS of $2.22 (+120% YoY), beating consensus ($92.3B revenue, $2.09 EPS). The company lifted AI/data-center performance with data center revenue of $89B (+117% YoY) and guided Q3 revenue to $108B (+89% YoY) with gross margin dipping slightly to ~74% at midpoint. Shares rose more than 4% after hours and the valuation returned to levels not seen since early 2019 (below 27x earnings), reinforced by an expanded AWS partnership to deploy 3M+ GPUs.

Analysis

The bigger signal is not this quarter’s beat; it is that AI infrastructure spending is still early enough to re-accelerate despite already huge bases. That keeps the whole semiconductor supply chain in a scarcity regime, where suppliers with true bottleneck exposure—HBM, advanced packaging, networking, and power delivery—can keep pricing power longer than the market expects, while second-tier GPU challengers face a tougher unit-economics battle.

The AWS tie-up is strategically important for AMZN, but the market should separate ecosystem validation from near-term P&L. More AI deployment at AWS likely means higher capex and depreciation before it means better FCF conversion, so AMZN may trade more on operating margin discipline than on topline enthusiasm over the next 1-3 quarters. That creates a subtle winner/loser split: NVDA monetizes the cycle now; hyperscalers monetize later and only if utilization stays high.

Contrarian risk is that investors may be underpricing a future “good enough” phase where custom silicon, export constraints, or internal cost pressure slow incremental wallet share gains. The key falsifier is not demand disappearing; it is gross margin reverting faster than expected or sequential guidance decelerating into the next print. If that happens, the multiple can compress even with strong absolute growth, because the market is paying for duration, not just earnings growth.

Key Decisions for Investors

• Long NVDA on any 1-2 day post-gap consolidation; target 10-15% upside over 1-3 months if the next guide confirms >80% y/y growth. Falsifier: sequential revenue deceleration or gross margin slipping below the mid-73s.
• Pair trade: long NVDA / short AMD for 1-3 months. Thesis: the market will continue rewarding the incumbent with the clearest supply/demand visibility while AMD still needs multiple quarters of proof on performance, software, and scale. Risk: an AMD design-win surprise or faster-than-expected share gains.
• Add AMZN on weakness, not strength, for a 6-12 month horizon. The AWS news supports strategic relevance, but the trade works only if investors stay focused on future AI utilization rather than near-term capex drag. Falsifier: AWS margin commentary weakens or FCF inflects lower for multiple quarters.
• Watch the AI infrastructure basket rather than chasing only NVDA: SOXX, SMH, or names like ANET, AVGO, and MU should continue to benefit if capex remains elevated. Use as a confirmation trade; if the basket stops following NVDA, that is an early sign the market is questioning spend durability.