Nvidia Just Did Something It's Never Done Before. Here's Why It Makes the Stock a Buy.

Nvidia reported fiscal Q2 revenue growth of 106% to $96.2B and adjusted EPS up 120% to $2.22, beating $92.3B revenue and $2.09 EPS expectations. Guidance was also a standout: fiscal Q3 revenue is projected around $108B (~89% growth) and the company forecast 70% revenue growth in fiscal 2028, citing supply constraints. Data center revenue rose 117% to $89B and the quarter produced $24.1B operating cash flow and $21.4B free cash flow, with the stock trading about 17x forward P/E on fiscal 2028 consensus.

Analysis

The key market implication is that NVDA is transitioning from a single-product demand story to a de facto infrastructure tax on AI capex. That matters because it raises the durability of spend across hyperscalers, neoclouds, and enterprise buyers, which should keep estimate revisions positive for several quarters and support a higher terminal multiple than a pure semiconductor name. The first-order winner is NVDA; the second-order winners are networking and rack-integrators with incremental attach, while GPU alternatives face a tougher share-acquisition path because platform breadth, not just chip performance, is becoming the moat.

The near-term risk is not demand, but digestion: when a stock trades on accelerating expectations, any hint of supply easing or moderation in customer capex can trigger multiple compression even if revenue still grows fast. Over 1-3 months, the most important falsifier is a guide that implies normalization in fiscal 2028 cadence or margin pressure from mix/supply constraints; over 6-18 months, the bear case is that frontier-model spend becomes less concentrated and pricing power leaks to system-level competitors. A more contrarian read is that the market is underappreciating how much of the upside is already in estimates, so the best risk/reward may be relative value rather than outright chasing the name.

Competitively, AMD and INTC remain structurally behind on ecosystem lock-in, while AVGO, ANET, and selected rack/power beneficiaries can still participate if AI buildouts broaden beyond GPUs. But if NVDA starts to internalize more of the stack, those secondary names could see less attach than the market expects. Watch for hyperscaler capex commentary, export-control headlines, and any evidence that lead times are shortening; those would be the earliest signals that the growth inflection is peaking.

Key Decisions for Investors

• Buy NVDA only on post-earnings consolidation or a 2-4% pullback; 3-6 month horizon. Risk/reward remains favorable if FY28 estimates ratchet higher, but trim if growth guidance stops inflecting or if the stock loses its post-print support level.
• Pair trade: long NVDA / short AMD for 1-3 months. The thesis is ecosystem breadth and revenue visibility versus execution risk for the challenger; cover if AMD shows unexpected design-win conversion or materially better AI server share.
• If seeking lower-beta exposure to the same capex cycle, express it via a basket long in AVGO/ANET over 6-12 months, but only after confirming order momentum; these names should benefit from networking and systems attach if NVDA-driven AI spend broadens.
• Set an alert for any hyperscaler capex slowdown or export-control escalation; either would be a catalyst to fade NVDA strength and could compress the multiple before the revenue story breaks.