Prediction: Sept. 10 Will Be a Big Day for Nvidia Shareholders. Here’s What to Watch.

Nvidia reported fiscal Q2 revenue of $96 billion, up 106% year over year, and profit of $59 billion, up 126%, driven by $89 billion in data-center revenue. At its Sept. 10 Goldman Sachs conference presentation, investors will focus on the Vera Rubin rollout—expected to contribute about 20% of Q3 data-center revenue—and possible details on its new stand-alone CPU opportunity. Nvidia forecasts 70% revenue growth in fiscal 2028, though tight memory supply, higher memory prices, and unresolved access to the China market remain key risks.

Analysis

The September 10 appearance is primarily a positioning event, not a fundamental reset: NVDA’s valuation now requires evidence that rack-scale adoption is expanding total wallet share rather than merely pulling forward accelerator purchases. The key incremental datapoint is the mix and availability of next-generation systems; a clean ramp would support gross-margin durability and reduce the bear case that each architecture transition creates a digestion quarter. Conversely, generic AI-demand commentary is unlikely to sustain a move because buy-side expectations already embed continued hyperscaler capex strength.

NVDA’s move into CPU-adjacent systems is more strategically important for AMD and INTC than for NVDA’s near-term P&L. Bundling compute, networking and software raises customer switching costs and can pressure AMD’s data-center attach rate, while INTC’s CPU exposure faces a further loss of relevance in AI-native clusters; neither outcome should be assumed from early product claims. The clearer second-order beneficiary is MU: constrained high-bandwidth-memory supply improves its mix and pricing, although a shortage severe enough to delay system shipments becomes a revenue-timing risk for NVDA and its server OEM ecosystem.

Consensus appears too focused on an upside soundbite and underweights two asymmetric risks over the next 1-3 months: component availability can convert demand into backlog rather than recognized revenue, and China optionality has low visibility despite any nominal policy reopening. For the next 6-18 months, the investable question is whether customers earn sufficient inference revenue to sustain capex after the initial training buildout; a deceleration in hyperscaler capex guidance, rather than a single NVDA conference comment, would be the meaningful thesis break.

Key Decisions for Investors

• Maintain, but do not add aggressively to, NVDA into September 10; use any event-driven 5-8% advance without a raised revenue or supply outlook to trim tactical exposure. Re-add only if management provides verifiable shipment/availability commentary supporting next-quarter revenue conversion; invalidate the constructive view on a material gross-margin guide-down or evidence of delayed system acceptance.
• Initiate a 1-3 month relative-value basket: long MU / short AMD, sized beta-neutral. Memory scarcity is directly monetizable for MU, while AMD faces incremental platform-bundling risk; target a 10-15% relative move, with a stop if MU signals HBM supply normalization or AMD demonstrates material AI-server share gains at its next earnings update.
• Keep INTC as an underweight or put-spread candidate rather than a direct pre-event short. AI-system CPU integration worsens its strategic narrative, but policy support and foundry headlines can dominate near-term trading; use 3-6 month downside structures only after confirming that data-center CPU guidance is not improving.
• Set an alert around hyperscaler capex guidance during the next earnings cycle. A broad reduction in AI infrastructure commitments is the higher-conviction catalyst to de-risk NVDA, MU and the semiconductor complex; absent that evidence, do not extrapolate a conference reaction into a structural demand inflection.