Nvidia Stock Is Below $200 Per Share Again. Here's When It Could Hit $300.

Nvidia is being pressured amid AI data-center overspending fears, but the article highlights management expectations that hyperscaler capex will exceed $1.0T in 2027 (from ~$650B in 2026). Analysts project EPS of $12.76 for fiscal 2028 (ending Jan 2028) and the piece values the stock at a 25x multiple to imply ~$319, suggesting potential upside of ~50% and a possible path to ~$300 in late 2027.

Analysis

The market is treating AI spend as a single-variable trade, but the more important mechanism is budget concentration: a few hyperscalers can keep capex elevated even while returns are still being debated, because falling behind on capacity is a worse failure mode than overspending. That supports NVDA’s revenue visibility over the next 1-3 quarters, but the stock still needs proof that supply stays tight and order books remain forward-dated; otherwise the multiple can compress faster than earnings grow.

Second-order winners are the bottlenecks around the bottleneck: HBM/memory, advanced packaging, optics, and networking tend to benefit before the broader AI software stack does. The losers are high-multiple software names that are being forced to fund AI infrastructure while monetization remains delayed; that crowding-out effect matters more than headline AI enthusiasm. If AI budgets keep shifting toward compute and interconnect, the relative performance gap between infrastructure and application-layer names should widen.

The real risk is not just a temporary capex pause, but share loss inside the AI dollar: custom ASICs, in-house accelerators, and inference optimization can preserve overall spending while quietly taking margin and mix away from NVDA over 6-18 months. Falsifiers are simple: any downward revision to 2026 hyperscaler capex, slower order cadence in the next earnings cycle, or a policy shock on export controls would argue the rebound thesis is too early. Conversely, if cloud capex prints keep rising, the selloff is likely more of a valuation reset than a fundamentals break.

Key Decisions for Investors

• Buy NVDA only on weakness after the next pullback, not on strength; target a 6-12 month horizon with upside tied to continued hyperscaler capex, and cut if cloud capex guides roll over or order lead times shorten.
• Pair trade: long NVDA / short IGV for 1-3 months to express the view that infrastructure spend continues while software monetization lags; thesis breaks if AI-app revenue conversion accelerates or software multiples de-rate less than expected.
• Use SMH as a lower-beta proxy if you want the AI infrastructure theme without single-name execution risk; this is cleaner if you believe the bottleneck is industry-wide rather than NVDA-specific.
• Set an alert on the next MSFT/AMZN/GOOGL/META capex commentary: a sustained uptick supports adding to NVDA, while any moderation should be treated as a signal to trim exposure.