Here's What a $10,000 Investment in Nvidia Stock Could Be Worth in 10 Years

The article projects Nvidia could reach roughly $1,200 per share by fiscal 2036, implying more than a fivefold gain from about $230 today, based on an assumed 15x forward P/E. Its model forecasts revenue rising to $4.62 trillion and EPS to $80.47 by fiscal 2036, supported by sustained AI-infrastructure demand, a gradual slowdown to 15% annual revenue growth, roughly 70% gross margins, and 2% annual share-count reduction through buybacks. The outlook is highly bullish but rests on aggressive long-term author assumptions rather than company guidance.

Analysis

The relevant debate is not whether NVDA remains the AI incumbent, but whether its earnings duration can outrun the inevitable shift in hyperscaler capex from training clusters toward lower-cost inference, custom ASICs and utilization optimization. The long-dated model embeds an unusually favorable combination of sustained high-teens-plus growth, limited pricing erosion and very high incremental margins; even modest deviations in any one of those variables would materially reduce terminal EPS and justify a lower multiple. The claimed inference expansion should be treated as unverified until transaction terms, customer migration and revenue contribution are disclosed.

Near term (days to 3 months), the article is unlikely to create a new catalyst; NVDA trades on hyperscaler capex read-throughs, supply availability, Blackwell/Rubin ramps and quarterly gross-margin guidance rather than retail valuation models. The more actionable second-order exposure is networking: AI cluster scale raises the strategic value of high-speed interconnect, supporting AVGO and ANET, while custom silicon adoption is a relative risk to NVDA but an opportunity for AVGO and MRVL over 6-18 months. Conversely, a broad AI spending deceleration would hit memory and foundry-linked names first through order digestion, even if NVDA maintains a larger share of a smaller spend pool.

Contrarian view: consensus may be underestimating NVDA's software and systems lock-in, but is also overextending current data-center economics into a decade-long terminal assumption. A move in NVDA should be sold only if evidence shifts from demand timing to demand destruction: hyperscaler capex cuts, falling accelerator lead times, a material gross-margin reset, or evidence that internally designed accelerators are displacing rather than supplementing NVIDIA deployments.

Key Decisions for Investors

• No event-driven add solely on this article. Maintain NVDA exposure only through the next earnings cycle if channel checks support accelerator lead times and hyperscaler capex guidance; trim if forward gross-margin guidance falls below consensus by more than 200 bps or if two major cloud customers reduce AI capex.
• For a 3-6 month relative-value expression, consider long NVDA / short SOXX in equal beta weights after broad semiconductor rallies. The thesis is that NVDA retains superior earnings visibility and pricing power while the index carries greater inventory and cyclical analog exposure; exit if NVDA revenue guidance misses consensus or SOXX earnings revisions broaden positively.
• For a 6-18 month diversification hedge against inference commoditization, build a measured long AVGO versus NVDA pair rather than a directional NVDA short. AVGO benefits if custom accelerators and networking take a greater share of AI budgets; risk is that merchant GPU demand remains capacity-constrained and custom programs slip, in which case close on deteriorating AI-semiconductor order commentary.
• Watch NFLX rather than act: it has no direct earnings linkage to the AI-infrastructure thesis. A material investment signal would require disclosed AI-driven content or advertising-margin savings, not generalized AI sentiment.