Nvidia Is One of the Most Profitable Businesses in the World. Here's Why the Stock Is Worth $311 a Share Right Now -- 39% More Than Its Share Price.

Nvidia is presented as undervalued despite a 64% net profit margin and average annual stock gains of 65% over the past decade. The article cites Nvidia's 25.1x forward P/E versus its five-year average of 34.4x, implying a potential value of about $311 per share, or 39% above the cited current price. The bullish thesis is supported by Nvidia's expanding AI data-center, software, networking, custom-processor, financing and Hugging Face platform initiatives, though the valuation estimate is explicitly assumption-dependent.

Analysis

The valuation argument is mechanically weak: Nvidia’s five-year average multiple spans a pre-AI earnings base and an extraordinary earnings step-up, so re-rating to that historical average is not a reliable upside framework. The investable question is whether FY27/FY28 data-center revenue and gross margin can remain above consensus while hyperscaler capex normalizes; a modest revenue miss or gross-margin guide-down would matter more than a low-20s forward P/E headline. Treat unverified strategic-expansion claims, including acquisitions and financing initiatives, as immaterial until disclosed in Nvidia filings or confirmed by counterparties.

Near term, positive retail commentary is unlikely to be a catalyst after NVDA’s broad institutional ownership and persistent AI narrative. Over 1-3 months, the relevant read-throughs are hyperscaler capex guidance, lead times for Blackwell/Rubin systems, HBM availability, and whether networking attach rates offset platform-transition costs. The second-order beneficiary of sustained accelerator demand is AVGO, where custom silicon and networking exposure provide a less concentrated route to AI spend; the principal competitive risk to NVDA is not a sudden GPU displacement but slower incremental unit growth as large customers internalize more workloads through ASICs.

The contrarian case is that NVDA may be correctly valued for a transition from scarcity pricing to a more normalized infrastructure cycle. A durable multiple expansion requires evidence that software, networking, enterprise inference, and financing offerings increase recurring revenue or reduce cyclicality; otherwise, these initiatives could dilute the simplicity of the high-margin accelerator model. Conversely, consensus may still underappreciate the earnings durability if inference demand broadens beyond the largest cloud buyers and installed-system upgrades remain on a three-year cadence.

Key Decisions for Investors

• Maintain, but do not add aggressively to, long NVDA ahead of the next earnings print; add only on a post-results decline if data-center revenue guidance is maintained or raised and gross-margin guidance remains within 150 bps of consensus. Upside is a renewed earnings-revision cycle over 1-3 months; thesis is falsified by hyperscaler capex cuts or a material margin reset.
• Express AI infrastructure exposure as long NVDA / short an equal beta-weighted semiconductor basket (SOXX) over the next quarter, rather than an outright momentum chase. This isolates Nvidia’s platform, software, and networking execution from broad rate-sensitive semiconductor multiple compression; exit if NVDA’s revenue-growth premium versus the basket narrows for two consecutive reporting periods.
• Watch AVGO as the cleaner incremental beneficiary if customer ASIC adoption accelerates. Initiate only after management confirms AI-networking or custom-silicon backlog conversion; the catalyst window is the next 1-2 earnings cycles, while risk is that ASIC growth cannibalizes merchant GPU spend faster than networking attach expands.
• Avoid using GOOG, NFLX, or FNMA as direct expressions of this item. GOOG is relevant only through cloud capex disclosure, while NFLX and FNMA have no sufficiently direct earnings sensitivity to Nvidia’s valuation or AI infrastructure demand.