Jensen Huang Told Investors in Seoul to "Buy at a Discount" During the Recent AI Stock Sell-Off. Here's Whether His Call Has Paid Off.

Jensen Huang’s June 8 “buy at a discount” comments have left Nvidia shares up ~2.2% since then, while the stock’s fiscal 2028 forward P/E barely changed (16.6x at the time vs ~16.5x by Aug. 21) as earnings expectations rose. First-quarter results showed $48.6B in free cash flow, but with $119B in manufacturing/supply/capacity commitments highlighting heavy build-out needs. With next results due Aug. 26 and fiscal Q2 revenue guided to $89.2B–$92.8B, the article frames Nvidia as reasonably valued yet still dependent on continued earnings growth.

Analysis

This is now an estimate-revision story more than a valuation story. The market is not paying a higher multiple for NVDA; it is accepting a higher earnings base, which means the stock remains vulnerable if the next print slows the pace of upward revisions. With consensus already leaning on persistent AI demand, the first real de-rating trigger is not a demand collapse but a guide that implies the street has moved too far, too fast.

Second-order, the real P&L pressure sits with the hyperscalers, not NVDA. AMZN, GOOGL, META, and MSFT can absorb capex for a while, but if AI monetization trails infrastructure spend, the hit shows up first in free-cash-flow yield, then in multiple compression, and only later in revenue. That creates a near-term asymmetry where NVDA can still outperform on operating leverage while the buyers of its chips face skepticism around capital intensity and depreciation drag over 1-3 quarters.

The contrarian risk is that the market is treating supply commitments as demand certainty. Commitments help NVDA visibility, but they are not the same as end-demand or margin durability; if customers slow orders or push out deployments, those commitments become a warning sign rather than a moat. The falsifier is a guide that fails to beat the low end of expectations, or commentary that indicates gross margin/FCF conversion is normalizing faster than estimates imply; that would likely compress the 2028 multiple even if the shares hold near-term support.

Key Decisions for Investors

• Do not add to NVDA into the Aug. 26 print; wait for the release. Buy only on a post-earnings pullback of ~5-7% if revenue and margin guide remain intact, with a clear stop if fiscal Q3 guidance comes in at or below the low end of consensus.
• For event-driven exposure, consider a limited-risk NVDA call spread expiring 4-8 weeks after earnings rather than stock outright; the setup works only if guidance re-accelerates estimate revisions, and the premium at risk should be capped.
• If earnings confirm AI demand but investors focus on capex payback risk, a relative-value short basket of AMZN/GOOGL/META/MSFT versus long NVDA is the cleaner expression for the next 1-3 months; the bet is that supplier economics stay stronger than customer economics while AI build-out remains spend-heavy.
• Set a watch level for NVDA around the earnings reaction: sustained trading below the pre-print range after guidance would signal that the estimate-upgrade cycle is stalling, which is the main invalidation point for the long thesis.