AI Buildout Hits Inflation as Fed Hikes Rates

Following Fed Chair Warsh's interest-rate increase, JPM Asset Management strategist Stephanie Aliaga said oil prices remain a long-term inflation driver while the AI buildout is also materially affecting the inflation and productivity outlook. She argued that a slower AI frontier could support broader adoption and ultimately lift productivity, creating a mixed macro backdrop of near-term inflation pressure and longer-term efficiency gains.

Analysis

The key transmission is not simply higher discount rates for megacap technology; it is the financing hurdle for the second tier of AI infrastructure. MSFT, GOOGL, AMZN and META can self-fund elevated capex, while levered data-center developers, power developers and smaller cloud/AI beneficiaries require a return on invested capital that may no longer clear at current power, equipment and debt costs. This favors hyperscalers over the broader AI-capex complex and should widen dispersion across VRT, ETN, PWR, CEG and private-data-center proxies over the next 1-3 quarters.

A moderation in frontier-model spending could be constructive for software adoption only if enterprise implementation converts into measurable labor or revenue gains. That creates a 6-18 month rotation opportunity from semiconductor-capex beta (NVDA, AMD, AVGO) toward application and workflow beneficiaries (MSFT, ORCL, CRM, NOW), but the near-term earnings risk remains that customers defer software seats while they reassess AI budgets. Productivity-led disinflation is a longer-duration thesis, not an immediate offset to a commodity-driven inflation impulse.

Oil is the critical falsifier for the duration trade: sustained higher energy prices raise inflation expectations while simultaneously squeezing consumer discretionary demand and transportation-intensive margins. The market may be underpricing this stagflationary combination, in which long-duration AI equities derate despite intact earnings, while XLE and select regulated power providers retain relative pricing power. A rapid decline in crude and a softer payroll/CPI sequence would reverse the relative-value setup quickly.

Key Decisions for Investors

• Initiate a 1-3 month relative-value position: long MSFT and GOOGL versus short an equal-dollar basket of VRT and ETN. Hyperscalers can absorb a higher cost of capital and may gain bargaining power if customer-funded data-center projects slow; exit if hyperscaler capex guidance is cut or 10-year Treasury yields fall materially below recent ranges.
• Reduce high-multiple AI infrastructure exposure into strength; specifically, use rallies in VRT, PWR and CEG to trim rather than add until project backlog conversion, utility interconnection timing and incremental funding costs are independently verified.
• Express inflation-tail protection through a modest long XLE versus short XLY position for the next 1-3 months. The trade benefits if energy remains firm and real household purchasing power weakens; stop out if crude breaks down decisively and inflation breakevens compress.
• For a 6-18 month watchlist, favor enterprise AI monetization beneficiaries NOW, ORCL and CRM over pure compute beta, but do not initiate aggressively until earnings calls demonstrate accelerating paid AI attach rates and stable net retention rather than pilot-program commentary.
• Monitor the 10-year yield, 5-year breakevens, crude prices and hyperscaler capex guidance as gating signals. A simultaneous decline in all three macro measures would falsify the defensive/relative-value thesis and reopen the case for broader AI infrastructure exposure.