Our newest stock seems unstoppable — plus, why Cramer calls Meta a 'screaming buy'

Chip stocks extended their rally for an 18th straight day, led by Intel's more than 20% surge on a strong quarter, while AI-linked names such as Arm, Corning, Microsoft, Meta and Amazon were highlighted favorably. Cramer said he remains constructive on Corning ahead of earnings and would not sell Microsoft, while calling Meta a "screaming buy" amid new CPU deals and layoffs. The broader market tone was mixed, however, with gains concentrated in tech and a lack of participation from other sectors.

Analysis

The key second-order signal is that capital is still concentrating in the AI supply chain rather than broadening into the rest of the market. That usually extends the winners faster than fundamentals alone would justify, because passive and quant flows chase the same scarce exposure, but it also leaves the tape vulnerable to a single-factor unwind if rates, oil, or one mega-cap guide even modestly lower. In the near term, the market is rewarding companies that can credibly show either capacity tightness or efficiency gains tied to AI infrastructure, which keeps the bar low for adjacent infrastructure names and high for broad industrial cyclicals.

Corning looks like the cleaner thematic expression than the hyperscalers because it sits closer to the bottleneck: if AI deployment remains power-constrained, optical connectivity becomes a forced purchase, not a discretionary one. The risk is not demand collapse but supply normalization — if hyperscalers announce incremental sourcing diversification or pricing eases, the multiple can compress quickly even while revenue still grows. That makes the next earnings print a volatility event: any commentary on order visibility or contract duration will matter more than headline EPS.

The more interesting contrarian read is that the “best” AI names may be becoming their own capital efficiency story. Meta’s combination of aggressive capex discipline and vendor substitution implies it can keep investing without letting margins reset materially, which supports the stock, but it also pressures component suppliers with less pricing power. Microsoft is the opposite: sentiment is already anticipating repair, so the setup is less about upside surprise and more about whether management can avoid another round of duration pressure from cloud deceleration or spend creep.

Intel’s move matters less as a standalone turnaround and more as a signal that the market is willing to underwrite any credible domestic semiconductor reset. If that enthusiasm persists, it can spill into second-order beneficiaries in packaging, foundry equipment, and substrate suppliers, but it also raises the probability of crowded longs and sharp mean reversion once the next earnings batch shifts focus from narrative to guidance.

Key Decisions for Investors

• Stay long GLW into earnings as a 1-4 week momentum/visibility trade, but size modestly and trim into strength; upside remains tied to contract commentary, while downside is a multiple reset if order growth appears less scarce than expected.
• Add on weakness to META versus the rest of mega-cap tech for a 1-3 month horizon; it has the best combination of capex discipline and operating leverage, but hedge with a short basket of lower-quality AI suppliers if margin compression broadens.
• Avoid chasing INTC after the gap move; only re-enter on a 10-15% pullback or on evidence of sustained design-win traction. The risk/reward is now skewed to post-event digestion rather than immediate follow-through.
• Pair long GLW / short a basket of weaker data-center hardware beneficiaries for the next earnings cycle; the thesis is that scarce fiber capacity is more defensible than adjacent components with easier substitution.
• Maintain a cautious underweight to MSFT for 2-6 weeks until guidance resets confidence; if the stock rallies on broad tech beta without confirmation of re-acceleration, use that strength to fade with tight risk controls.