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.

Nvidia shares are up about 2.2% since CEO Jensen Huang told reporters on June 8 to view the June tech sell-off as a “buying opportunity.” Valuation has stayed roughly stable as Wall Street lifted earnings expectations: fiscal 2028 forward P/E is about 16.5x (vs ~16.6x at the June 8 open), while free cash flow was $48.6B in fiscal Q1 alongside $119B of manufacturing/supply/capacity commitments. The near-term test is the Aug. 26 fiscal Q2 release, with revenue guided to $89.2B–$92.8B.

Analysis

The key market mechanism is not “NVDA is expensive,” but that the stock is now being priced as a near-perfect passthrough on hyperscaler capex. That creates a narrower path to multiple expansion: upside needs another revision wave in FY28 EPS, while a merely in-line print is enough to trigger de-rating because the valuation anchor has already shifted up with estimates. The risk is asymmetric around the next earnings release: before the guide, shares are being supported by estimate momentum; after the guide, the market will test whether that momentum is still self-funding or just front-loaded demand.

Second-order, the capex burden is increasingly landing on AMZN, MSFT, META, and GOOGL balance sheets and free cash flow statements. MSFT and AMZN are best positioned to monetize AI spend through cloud attach and enterprise workflow lock-in; META is the most exposed to capex intensity versus near-term monetization, so any disappointment in ad or engagement monetization would hit it harder than peers. GOOGL sits between the two: if inference demand stays strong, it can absorb capex better than the market expects; if not, AI spend becomes a margin drag rather than a growth driver.

The contrarian miss is that the “early innings” narrative may already be fully reflected in customer budgets, while supply commitments at NVDA imply revenue durability that still has to be proven. If hyperscalers slow incremental orders or if gross margin slips from mix/supply concessions, the stock can underperform even with healthy absolute growth. Falsifiers: a guidance raise with better-than-feared margin, or a clear signal that 2027 demand visibility is still tightening; downside catalysts are a flat guide, softer backlog commentary, or any evidence that customer capex is normalizing.

For now this looks more like a hold-the-core than a chase-the-rally setup. The cleanest trade is relative value: NVDA remains the best pure AI monetization vehicle, but the market is increasingly paying for that certainty with less room for disappointment.

Key Decisions for Investors

• Maintain / add to NVDA on post-earnings weakness only if Aug. 26 guide confirms revenue acceleration into FY27-FY28; upside case is another estimate revision cycle, while the main risk is a de-rating on merely in-line guidance.
• Pair trade: long NVDA / short META into the next 1-3 months if you want AI exposure with lower balance-sheet sensitivity on the long leg; META is the most likely hyperscaler to absorb AI capex without equivalent near-term monetization.
• Relative-value long MSFT vs. GOOGL over 3-6 months: MSFT has the cleaner path to monetizing AI spend through cloud and enterprise software attach, while GOOGL is more exposed to capex drag if ad monetization doesn't keep pace.
• Set an alert on NVDA FY27/FY28 EPS revisions and gross margin commentary after earnings; if estimates stop rising while the stock remains near current levels, the multiple support case weakens materially.
• Avoid forcing a bullish options expression pre-earnings unless implied volatility is clearly mispriced; the event risk is binary and the better entry is likely after the print if guidance is strong.