AMD Stock and Intel Crushed Nvidia in the First Half. Here's My Prediction for the Second Half.

In 1H, investors rotated into AMD and Intel, sending AMD up 171% and Intel up 278% vs Nvidia’s ~7.2% gain, but the article argues Nvidia can reassert leadership in data-center CPUs in 2H. It cites Nvidia’s plan to launch its first stand-alone CPU (Vera Rubin) and targets $20B in stand-alone CPU sales this year, alongside the view that CPUs are key for agentic AI. Nvidia also appears cheaper at ~22x forward earnings, while Intel reported 6 straight quarters of revenue beats (revenue +7% in the latest quarter), supporting the bullish case that Nvidia could outperform AMD/Intel from here.

Analysis

The important mechanism is not that NVDA is “entering CPUs,” but that it is trying to move from being a premium accelerator supplier to owning more of the AI system stack. If it can bundle CPU + GPU + interconnect into one procurement decision, it raises switching costs for hyperscalers and puts pressure on AMD/INTC to compete on price in the one segment where margins are already most vulnerable: data-center sockets.

Near term, this is mostly a sentiment and positioning catalyst; the real financial impact will lag by multiple quarters because qualification cycles for server platforms are slow. The market is likely to overestimate first-wave adoption and underestimate how much of the CPU market is constrained by software compatibility, platform inertia, and customer preference for dual-sourcing. That means the trade works best if NVDA can show visible design-win momentum into the fall product cycle; without that, the move can fade quickly.

Contrarian view: consensus is treating the CPU roadmap as a new growth leg, but the more durable implication may simply be moat defense. NVDA does not need to “win” most CPUs for this to matter; it only needs enough attach rate to keep AI buyers inside its ecosystem and prevent AMD from taking incremental wallet share. The biggest loser is probably INTC if AI server procurement shifts toward integrated platforms, while AMD is the higher-quality competitor and therefore the less attractive outright short.

Key Decisions for Investors

• Long NVDA / short INTC as a 3-6 month relative-value trade: INTC has the most execution risk if the AI-CPU narrative stalls, while NVDA gets the highest multiple support from any credible platform expansion.
• If initiating new exposure, prefer a NVDA call spread into the fall launch window rather than stock: upside is tied to product/attach-rate surprises, but the event path is binary enough to justify capped-premium risk.
• Avoid chasing AMD on the same headline; use rallies to trim. AMD is better positioned fundamentally than INTC, so it is the weaker short, but its recent rerating leaves less room for disappointment on valuation compression alone.
• Set a watch item on NVDA’s next earnings for explicit CPU attach-rate commentary and any customer design-win disclosures; absent that, treat the story as narrative-driven and reduce exposure if post-launch adoption signals are not visible.
• Falsifier for the bullish NVDA-relative thesis: if Rubin/CPU rollout slips, or if hyperscaler capex commentary shows no incremental platform adoption by the next 1-2 quarters, rotate back toward sector-neutral exposure.