Wall Street sees 'changing of the guard in AI' as Intel, AMD shares soar while Nvidia lags

AI infrastructure spending is broadening beyond Nvidia, with AMD, Intel, Micron and Corning all posting outsized gains as investors rotate into CPUs, memory and optical components. Micron topped an $800 billion market cap and is up more than 750% over the past year, while AMD's results and guidance beat estimates and Intel rallied on Apple-related chip reports. The article also highlights a potential 25% to 30% correction risk in semis after a 66% YTD surge in the PHLX Semiconductor Index.

Analysis

The market is starting to price AI as a multi-node industrial cycle rather than a single-platform GPU story. That is a better setup for the hardware stack breadth, but it also means the marginal winner is moving from pure compute to bottleneck components with the tightest capacity and highest switching costs: memory, advanced interconnect, packaging, and server CPUs. In practice, that broadens earnings power outside NVDA, but it also raises the probability that the fastest-moving names are the ones with the most cyclical upside and the most abrupt air pockets if supply normalizes.

Memory looks like the cleanest near-term squeeze because the supply response is slow and customer behavior is highly inelastic once system buildouts are committed. The second-order effect is that OEMs and hyperscalers will likely ration inventory by mix, favoring higher-margin AI server configs over broad-based deployment, which can delay volume but intensify pricing power for the suppliers still in shortage. That is bullish for the current winners, but it also sets up a classic late-cycle dynamic: the best stocks are often the first to peak when lead times begin to stabilize.

CPU strength is more interesting as a signal than as a one-quarter trade. The AI narrative is shifting from training to inference and agentic workloads, which increases the importance of general-purpose processors, but this also invites faster competitive response from incumbent platform owners and could compress long-run share gains if hyperscalers optimize more aggressively in-house. Intel’s move is more of a sentiment reset than a fundamental clean-room turnaround; any deal flow headlines can move the stock sharply, but execution risk remains high over a 6-12 month horizon.

The contrarian read is that the market may be underestimating how much of this is a capex air-pocket trade rather than a durable re-rating of end-demand. If hyperscaler budgets re-phase, the most levered beneficiaries can correct 20-30% quickly even while the secular AI story stays intact. The key tell will be whether supply constraints convert into visible backlog and pricing discipline in upcoming guidance, or whether the current surge reflects positioning chasing a narrow set of shortage headlines.

Key Decisions for Investors

• Long GLW vs short a basket of lower-quality industrial tech suppliers for 1-3 months: use GLW as the cleaner optical bottleneck beneficiary with better visibility; downside is if Nvidia’s fiber transition slips or factories get pushed out, but the asymmetry still favors continuation on contracted demand.
• Buy AMD on any 3-5% pullback over the next 2-4 weeks; target a 15-20% upside continuation if AI server CPU share gains persist. Stop if hyperscaler commentary turns to inventory digestion or custom silicon displacement.
• Express a tactical long in MU for 4-8 weeks, but size it as a momentum trade, not an investment. Risk/reward is attractive while shortage pricing remains intact; take profits into any sign of lead-time normalization or commentary about customer allocation easing.
• Avoid chasing INTC after the recent vertical move; use call spreads instead of stock if participating over 1-2 months. The setup is headline-sensitive and prone to 15-20% air pockets if any Apple-related deal timing or scope is less than implied.
• Reduce or hedge NVDA relative to a basket long of AMD/MU/GLW over the next quarter. The thesis is not bearish NVDA fundamentals, but that the market is rotating toward second-order beneficiaries with more incremental upside as the AI stack broadens.