Jensen Huang Just Delivered Great News to Nvidia Shareholders

Nvidia reported fiscal 2027 Q2 adjusted EPS of $2.22 (+120% YoY) on revenue of $96.22B (+106% YoY), topping Wall Street expectations ($2.10 EPS; $92.17B revenue). The company guided Q3 revenue to $108B (+/−2%, ex China data center revenue) and gross margin to 74% (+/−0.5%), ahead of the $104.2B consensus, and repurchased $26B of stock with $99B remaining under authorization. CEO Jensen Huang said AI has reached an "inflection point" as demand accelerates. Shares were up about 3.8% into/after the release.

Analysis

This reads as a confirmation that AI spend is still broadening, not just rotating among a few hyperscalers. The first-order winner is NVDA, but the second-order winners are the bottlenecked parts of the stack: HBM memory, advanced packaging, and foundry capacity. That matters because sustained volume plus resilient margins implies the constraint is still supply, which usually supports the whole semiconductor complex until capacity catches up.

AMD and INTC face a tougher relative bar from here. A beat from the category leader with stronger forward economics makes it harder for competitors to justify premium multiple expansion unless they show a credible inflection in accelerator or AI CPU share. Near term, the sector can still squeeze higher, but over 1-3 months the market will start discriminating between companies with actual AI monetization and those with roadmap promises.

The contrarian risk is that the market treats this as proof of a multi-year supercycle when it may simply be a very strong phase of customer pull-forward. If hyperscaler capex growth slows into the next reporting cycle, the multiple can compress even if absolute revenue stays elevated. The key falsifier is any sign that forward margin guidance rolls over or that incremental revenue is becoming too concentrated in a few customers.

The buyback is also important: it reduces float at a time when positioning is crowded, which can extend the squeeze, but it does not eliminate valuation risk if growth normalizes. The right lens is not 'can NVDA grow more' but 'can competitors close the gap fast enough to matter to the market.'

Key Decisions for Investors

• Long NVDA on any 3-5% post-earnings pullback over the next 1-2 weeks; target 10-15% over 1-3 months if guidance holds and gross margin stays in the mid-70s. Falsify if management has to walk down margin or China-ex-China demand weakens.
• Pair trade: long NVDA / short AMD for 1-3 months. Thesis is continued share dominance at the high end while AMD needs several quarters of proof to re-rate; risk is a broad AI-semiconductor squeeze that lifts both legs, so keep gross exposure modest.
• Alternative pair: long NVDA / short INTC into the next earnings season. INTC is more exposed to the 'AI CPU' narrative gap; this works only if the market keeps rewarding actual revenue conversion over roadmap announcements.
• If already long broad tech, consider rotating incremental capital from QQQ into SMH for the next quarter. This captures AI capex upside with less hyperscaler concentration risk; invalidate if semis fail to outperform QQQ on the next downside tape.