AI spending isn’t the only thing from keeping the US economy from falling off a cliff

The article argues that AI spending is helping keep U.S. growth around 2%, with payrolls up 178,000 in March, unemployment claims below 200,000, and business investment in the AI build-out rising more than 10%. It says inflation remains elevated at 3.5% headline and 3.2% core, but frames the Iran conflict, higher gasoline prices, and tariffs as temporary headwinds rather than structural deterioration. Overall, the piece is constructive on growth and AI-driven productivity despite near-term inflation and geopolitical noise.

Analysis

The market is still underestimating how much of today’s nominal growth is being subsidized by capex intensity rather than end-demand. That matters because AI infrastructure is not just a “theme”; it is pulling forward orders across semis, power, networking, cooling, and industrial automation, which tends to create a second-wave earnings upgrade cycle even after the initial build phase slows. The bigger implication is that a softer labor market is not automatically bearish if productivity gains start showing up in margins before headcount meaningfully rolls over.

The more interesting trade is not whether AI spending peaks, but whether the post-build productivity step-up arrives before policy or geopolitics tightens financial conditions. If that happens, the market can sustain high multiples longer than bears expect because earnings breadth improves while inflation stays contained enough for rate cuts to remain on the table. The key risk is that energy-driven inflation reaccelerates just as capex starts to decelerate, which would hit consumers, compress multiples, and expose cyclical areas that have been riding fiscal and AI momentum.

Consensus is likely too linear on both the AI bubble and the war shock. The bear case assumes capex stops and the economy drops into a void; the bull case assumes productivity offsets everything immediately. The more plausible path is lumpy: a few quarters of “bad” headline employment but decent corporate profits and capex-led growth, followed by a rotation from builders to beneficiaries. That favors names with direct leverage to AI monetization or to lower input costs from productivity gains, rather than purely speculative AI infrastructure proxies.

Key Decisions for Investors

• Long MSFT / NVDA on a 3-6 month horizon: pair the software monetization leader with the hardware enabler; if AI spend broadens into recurring workflow adoption, the combo should outperform the market’s current ‘infrastructure-only’ framing. Use a 10-15% trailing stop if capex guide-downs appear.
• Initiate a relative-value long XLI / short XLY basket: industrials with automation, electrification, and data-center exposure should benefit from productivity-led margin expansion, while consumer discretionary remains more vulnerable to energy-driven real income pressure. Best entered on any CPI or gasoline spike-induced selloff.
• Buy 6-9 month calls on AMZN or GOOGL with strikes ~8-12% above spot: both can monetize AI through ads/cloud without requiring perfect unit economics, and the market still underprices the earnings leverage if productivity gains start showing up in operating margins.
• Fade speculative AI software names with no revenue acceleration via short baskets or put spreads: if the market begins rewarding actual monetization over narrative, the weakest balance-sheet names are most exposed to a 20-30% drawdown in 1-2 quarters.
• Maintain a tactical long in F as a lagging beneficiary only if fuel prices stabilize; if oil remains elevated, consumer affordability and lease/financing sensitivity cap upside, so treat it as a trade not a structural long.