Here's Why Nvidia and Broadcom Are Still Leading the Pack for AI Investing

Nvidia reported latest-quarter revenue of $68.1 billion, up 73% year over year, while analysts expect growth to accelerate to 79% next quarter and 85% the quarter after. Broadcom's fiscal Q1 AI semiconductor revenue rose 106% year over year to $8.4 billion, underscoring strong demand for custom AI chips. The article argues both stocks remain attractive long-term AI beneficiaries despite already strong share performance.

Analysis

The market is increasingly rewarding the companies that monetize AI demand directly rather than the broader ecosystem that merely enables it. That matters because the next leg is likely a capital-allocation race: hyperscalers will keep funneling spend toward compute where utilization is highest, which should preserve pricing power for the two dominant chip architectures while squeezing smaller accelerator vendors and slower-moving foundry capacity. The second-order winner is the semiconductor equipment and advanced packaging stack, where bottlenecks can keep margins elevated even if end-demand eventually normalizes.

The key risk is not demand collapse, but digestion. If enterprise AI budgets get rephased over the next 1-2 quarters, the stocks most exposed to “perfect execution” can de-rate quickly even while fundamentals remain strong, because expectations are already embedded in estimates. A more subtle hazard is customer concentration: when a handful of buyers account for most incremental orders, any internal design win or insourcing decision can shift revenue mix abruptly over a 12-24 month window.

The consensus is probably underestimating how durable the custom-chip trend is versus the market’s habit of treating it as a niche complement to general-purpose GPUs. In reality, custom silicon is an adoption unlock for workloads where power, latency, and unit economics matter more than flexibility, which means the total addressable market expands rather than cannibalizes outright. That said, the upside is likely to be more durable for the platform leader with ecosystem lock-in, while the custom silicon beneficiary may see lumpier but higher incremental surprise.

INTC is the quiet loser in this frame: even absent direct mention, any improvement in AI compute efficiency raises the bar for legacy x86 positioning unless it can capture adjacent packaging, networking, or design-win share. If AI infrastructure spending broadens beyond a few flagship customers, the supply chain beneficiaries should outperform on revisions, not just headline revenue growth.

Key Decisions for Investors

• Maintain core long NVDA into the next 1-2 earnings cycles; risk/reward remains favorable as long as estimate revisions stay upward and supply constraints persist. Use 10-15% trailing stops rather than fixed price targets given elevated multiple sensitivity.
• Add AVGO on pullbacks for a 6-12 month hold; custom silicon is a slower-burn monetization curve with fewer consensus owners. Best entry is after any post-print digestion, when implied growth expectations reset but design-win visibility remains intact.
• Pair long NVDA / short INTC over the next 3-6 months to express the widening gap between AI monetization leaders and legacy compute exposure. This is a cleaner relative-value trade than outright shorting semis because it isolates share gain versus secular relevance.
• Buy NVDA or AVGO call spreads 2-4 months out ahead of earnings when implied volatility is reasonable; the setup favors upside surprise from supply ramp or order acceleration, but spreads cap premium bleed if the market pauses.
• Monitor equipment and advanced packaging as secondary longs on any AI capex broadening; if shipment lead times extend further, the best risk/reward may come from the bottleneck, not the chip designer.