Big Tech earnings show how big, smart spending can be rewarded by the market

The article argues that AI-related capex is proving justified, citing Alphabet capex of $180B-$190B with the stock up 12% weekly, Amazon at $200B with AWS growing 28% and shares up 1.6%, while Apple rose 3.4% and Microsoft and Meta fell 2.4% and 9.8%, respectively. It highlights strong cloud growth at Google Cloud (63% growth, $20B quarterly revenue) and AWS ($37.6B quarterly revenue), while framing Microsoft and Meta as more vulnerable due to weaker monetization and heavier spending. Overall, the piece is bullish on AI infrastructure spend and constructive on Alphabet, Amazon, and Apple, while cautious on Microsoft and Meta.

Analysis

The market is no longer rewarding “AI spend” as a single factor; it is now pricing the quality of monetization behind the spend. The winners are hyperscalers that convert capex into visible marginal revenue today, while the market is penalizing firms whose capex is still mostly an option on future product relevance. That creates a powerful second-order effect: capital is likely to keep migrating away from generalized software/platform stories and toward infrastructure names with direct exposure to compute, power, cooling, networking, and custom silicon.

The most interesting implication is that this is not just a semis trade, it is a resource-allocation trade. If compute remains the binding constraint, the bottleneck shifts to electricity, interconnect, liquid cooling, and backup power, which supports earnings durability for the picks-and-shovels complex even if headline AI enthusiasm cools. The more these buyers accelerate, the more vendors with design wins gain pricing power and backlog visibility, while weaker software franchises face a rising bar for retaining multiples.

The risk is that this becomes a narrow leadership trade rather than a broad capex boom. If hyperscaler returns on capital take longer to show up, the market can quickly re-rate the “good spend” names lower too, especially if capex remains elevated into the next guide cycle without corresponding margin expansion. Over the next 1-3 quarters, the key reversal trigger is not lower AI demand; it is any sign that incremental spend is no longer translating into measurable revenue acceleration or operating leverage.

Contrarianly, the consensus may be underestimating how much of the upside is already embedded in the supply chain winners versus the headline platforms. The bigger dislocation may be in names that are still priced as cyclical industrials or component suppliers despite having secular order books tied to multi-year buildouts. Conversely, the market may be underestimating the downside in enterprises where AI is likely to compress rather than expand software monetization per user.

Key Decisions for Investors

• Overweight GOOGL and AMZN versus MSFT/META for the next 3-6 months; the former have clearer near-term monetization of capex, while the latter face higher odds of multiple compression if spend keeps rising faster than revenue conversion.
• Initiate a long ANET / short CSCO pair trade over 1-2 quarters; AI networking growth should continue to outpace legacy enterprise networking as new clusters demand higher-performance fabrics and lower-latency switching.
• Add to ETN and GEV on any 5-10% pullback; power and grid names have the cleanest second-order exposure to sustained AI buildout and can rerate as backlog converts into visible EPS growth over the next 2-4 quarters.
• Use call spreads in CRWV or NBIS for a tactical 3-6 month upside expression; these names should benefit if compute scarcity remains acute, but size modestly because valuations can de-rate quickly if hyperscaler demand normalizes.
• Avoid fresh longs in MSFT and META until the market sees evidence that incremental capex is translating into durable incremental revenue; treat any bounce as an opportunity to reduce exposure rather than chase.