Nvidia Would Have to Add Almost a Whole Broadcom to Reach $7 Trillion

Nvidia, valued at roughly $5.45 trillion, could reach $7 trillion—implying a 28% rise to about $290 per share—if it delivers its forecast growth and trades at an approximately 18x fiscal-2028 forward P/E, near Broadcom's current multiple. The company reported fiscal Q2 2027 revenue of $96.2 billion (+106% YoY) and net income of $59.7 billion (+126%), while forecasting about 70% revenue growth in fiscal 2028, potentially taking sales to roughly $700 billion. The principal risk is valuation: Nvidia's forward P/E has been compressing despite earnings growth, and continued multiple contraction could prevent the stock from reaching the $7 trillion milestone on the projected timeline.

Analysis

The relevant debate is no longer whether NVDA can grow earnings, but whether AI infrastructure spending can remain sufficiently concentrated in merchant GPUs to support its terminal multiple. A supply-constrained order book can mask demand elasticity: hyperscalers are committing capital before enterprise AI monetization is proven, raising the risk that 2027-28 procurement becomes a digestion cycle rather than a linear extension of current run-rates. The first sign of impairment would likely be customer capex guidance or lead-time normalization, not an abrupt NVDA revenue miss.

Margin durability is the more underappreciated variable. A 200-300bp gross-margin reset is manageable in isolation, but at NVDA's scale it can coincide with mix pressure from networking, rack-scale systems, and customer-specific configurations, reducing the premium multiple investors assign to the business. Meanwhile, supply expansion gives customers more negotiating leverage and accelerates the strategic rationale for internally designed accelerators; AVGO is a cleaner public proxy for that substitution than a directional semiconductor basket.

Near term, this type of milestone-oriented retail narrative is not itself a catalyst and may reinforce crowded positioning into earnings. Over 1-3 months, the key upside catalyst is evidence that next-generation system ramps are being pulled forward without incremental margin concessions. Over 6-18 months, the central risk is not low-end demand but a lower valuation regime if returns on hyperscaler AI capex lag depreciation and power costs.

Contrarian view: consensus treats a lower earnings multiple as an exogenous valuation problem. It may instead be the market discounting a shift from scarcity rents toward a more competitive, systems-integration-heavy industry structure. That argues for owning AI compute exposure selectively rather than assuming all incremental AI spend accrues to NVDA.

Key Decisions for Investors

• Maintain NVDA as a core but do not add solely on the valuation milestone narrative; add only after the next earnings call if backlog conversion, customer concentration, and gross-margin guidance support stable-to-improving FY2028 earnings revisions. Thesis is falsified by a material lead-time reduction, sub-72% normalized gross-margin outlook, or hyperscaler capex cuts.
• Initiate a 6-12 month relative-value position: long AVGO / short NVDA, sized beta-neutral, if the ratio approaches recent lows. The trade monetizes custom-ASIC share gains and reduces exposure to a broad AI capex drawdown; stop if NVDA delivers accelerating networking/compute growth while AVGO's custom-silicon commentary fails to show program ramps.
• For directional NVDA exposure, prefer a 3-6 month call spread funded with an out-of-the-money put spread rather than unhedged stock at elevated positioning. Target upside only through the next earnings revision cycle; cap risk if the stock breaks below the post-earnings support level on unchanged fundamentals.
• Monitor MSFT, AMZN, GOOGL and META capex guidance, TSM capacity commentary, and NVDA lead times as weekly thesis indicators. Two or more hyperscalers signaling AI-capex optimization rather than expansion would warrant reducing semiconductor beta before reported demand weakens.