2 Brilliant Industrial Stocks to Buy for the AI Spending Boom

Caterpillar and Vertiv are highlighted as AI data center beneficiaries, with Caterpillar's power and energy division revenue up 21% to $7.0B and management now expecting power generation product sales to triple by 2030 versus 2024. Vertiv posted first-quarter net sales of $2.65B, up 30% year over year, with non-GAAP free cash flow rising 147% to $653M and guidance lifted to 2025 revenue of $13.5B-$14.0B and adjusted EPS of $6.30-$6.40. The piece is bullish on both companies' AI-linked growth outlook, especially Vertiv.

Analysis

The real trade here is not generic “AI beneficiaries,” but bottlenecks in power delivery and onsite generation. The second-order winner set is broader than CAT and VRT: electrical contractors, switchgear vendors, gas turbine/service names, and grid interconnect players should see a multi-quarter uplift as customers pay up to bypass utility queues. The loser set is subtler — hyperscalers and colo operators with weaker balance sheets will be forced to accelerate capex, pressuring free cash flow near term even as long-term revenue visibility improves.

CAT is more of a durable, lower-beta way to express the data-center buildout than a pure AI torque name. The market may still underappreciate that primary power generation for AI sites can resemble a multi-year installed-base upgrade cycle rather than a one-time equipment sale, which supports aftermarket/service revenue and margin durability. The risk is that some of the order strength is pull-forward: if grid approvals, gas availability, or financing costs improve in 2H25-26, incremental demand could normalize faster than consensus expects.

VRT is the cleaner growth expression, but also the more vulnerable to multiple compression if guidance keeps rising slower than the stock has already priced in. The key debate is whether liquid cooling and critical power remain capacity-constrained enough to sustain >25% growth for several more quarters; if supply catches up, mix and pricing can fade quickly. Consensus is likely missing how fast competitors will chase the margin pool, which could turn a currently scarce franchise into a more competitive install-base contest by 2026.

The contrarian view is that the AI infrastructure spend cycle is still under-penalized for timing risk, not direction risk. That argues for owning the picks-and-shovels, but doing so with discipline: the upside is a multi-year capex supercycle, while the downside is a valuation reset if order growth decelerates before revenue catches through. In other words, this is a good theme, but the trade should be structured around earnings revisions, not narrative momentum.

Key Decisions for Investors

• Go long CAT on a 3-6 month horizon as the lower-volatility AI infrastructure proxy; target a modest multiple expansion if power-generation guidance keeps inflecting, but use a tight stop if order growth slows for two quarters.
• Own VRT only on pullbacks or via call spreads rather than outright stock; reward/risk is attractive if it keeps beating guidance, but the current setup is vulnerable to 15-20% drawdowns on any guidance pause.
• Pair trade: long CAT / short a weaker electrical-equipment or thermal-management peer that is more exposed to single-end-market concentration; the thesis is relative share gain as customers prioritize proven uptime vendors.
• Add a basket long of AI infrastructure enablers with CAT + VRT + select switchgear/generator names over a 6-12 month window; this captures the capex cycle while diversifying idiosyncratic execution risk.
• If VRT gaps higher on another guidance raise, consider selling upside calls into strength; the stock can keep working, but implied expectations are likely to outpace near-term fundamental upside.