AI Spending Is Nearing $1 Trillion Annually. These 2 Stocks Are Most at Risk If It Slows.

Nvidia reported fiscal 2027 Q2 revenue of $96.2 billion, including $89 billion from data centers, while Micron reported fiscal 2026 Q4 revenue of $54.2 billion and full-year revenue of $133.2 billion, both records. The article highlights their exposure to continued AI spending, with Nvidia also facing competition and export restrictions, and Micron vulnerable to a memory supply glut and falling margins. Analysts’ average 12-month targets imply roughly 51% upside for Nvidia and 75% for Micron, but the article cautions that a slowdown in AI investment could pressure both stocks.

Analysis

The key distinction is demand visibility versus cycle exposure. A pullback in hyperscaler capex would hit NVIDIA through delayed GPU orders and weaker expectations for future growth; substitution by custom silicon or more efficient inference could compound that effect even if total AI workloads keep rising. Advanced Micro Devices may capture some displaced accelerator demand, but only if it demonstrates competitive performance, software readiness, and supply—not merely because buyers diversify.

Micron has a different downside: memory pricing and utilization can weaken as new capacity arrives, potentially before end-demand contracts. HBM qualification and customer concentration may support relative resilience, but do not immunize the broader memory business from a pricing cycle. Watch memory contract pricing, inventory, and capacity plans alongside hyperscaler capex guidance.

Near term, guidance and order commentary matter more than backward-looking records. Over 1–3 months, capex revisions and memory pricing are the most useful catalysts; over 6–18 months, capacity additions and custom-chip adoption determine whether today’s economics persist. A contrarian risk is that investors treat AI spending as one aggregate: workload growth can continue while GPU intensity or memory pricing falls. Separately, the article’s Micron revenue figures appear internally inconsistent; verify against company filings before using them in any model.

Key Decisions for Investors

• Prefer NVIDIA over Micron on a relative basis if positioning for an AI-spending slowdown: consider a paired long NVDA / short MU, sized for the possibility that both decline. Reassess if Micron’s filings and forward guidance confirm durable HBM growth alongside stable broader memory pricing.
• Avoid treating AMD as an automatic hedge to NVIDIA. Use it as a watchlist beneficiary only if product adoption, software traction, and supply availability show measurable share gains.
• Track hyperscaler capex guidance and order timing for the 1–3 month demand signal; track memory contract prices, customer inventories, and producer capacity plans for the 6–18 month cycle signal. Reduce the relative trade if capex accelerates and memory pricing remains firm.
• Before updating Micron estimates, reconcile the article’s revenue claims with reported filings. Falsifiers for the cautious thesis include sustained memory-price strength despite capacity additions, or NVIDIA order/guidance resilience through customer capex moderation.