
Caterpillar and Eaton are highlighted as AI infrastructure beneficiaries, with Caterpillar reporting record quarterly sales of $19.1B and a $51B backlog, while Eaton posted record adjusted EPS of $3.33 and all-time high quarterly sales of $7.1B. Caterpillar’s Power & Energy sales rose 23% to $9.4B and Power Generation sales jumped 44% YoY, underscoring data center demand; Eaton’s free cash flow increased 17% YoY to a company-high $1.6B. The article also emphasizes shareholder returns, including Caterpillar’s $7.9B in buybacks and dividends in FY25 and Eaton’s 7.8% five-year dividend growth rate.
The market is starting to price AI not as a semiconductor-only story but as a multi-year utility buildout, and that’s the key second-order shift. CAT and ETN sit in the least glamorous but most durable part of the stack: power generation, distribution, and thermal management. That matters because the bottleneck is no longer just compute demand; it is grid interconnection, onsite generation, and electrical redundancy, which tends to support longer order visibility and less cyclicality than the market typically assigns to industrials. The more interesting implication is competitive pressure on adjacent infrastructure vendors. As hyperscalers push for faster deployment, spend should cascade into switchgear, transformers, breakers, controls, and large-engine service demand, while pure software beneficiaries get less of the incremental capital intensity than headlines imply. Suppliers with long-cycle, capacity-constrained manufacturing footprints should retain pricing power for several quarters, but that also raises the probability of bottlenecks and margin surprises elsewhere in the electrical supply chain. The contrarian risk is that this becomes a consensus overcrowding trade in "AI power" and the market front-runs several years of growth into a shorter window. If data center permitting, interconnect delays, or utility capex execution slips, order growth could remain strong while shipment timing shifts right, creating a multiple compression event despite solid fundamentals. A sharper risk is policy or procurement normalization: if hyperscalers phase spending after a rush of buildouts, these names could de-rate faster than earnings estimates roll over because the valuation is already discounting an extended shortage regime. The most actionable setup is to stay long the infrastructure enablers but not as a blanket basket. CAT has better asymmetry if power-generation demand broadens beyond data centers into backup and grid resilience, while ETN is more directly levered to the electrical bottleneck and could sustain a higher multiple if backlog converts cleanly; the better trade is owning both versus chasing the more crowded AI beneficiaries. I would avoid extrapolating this into a broad industrial bull thesis until we see whether backlog growth keeps accelerating into the next two quarters.
AI-powered research, real-time alerts, and portfolio analytics for institutional investors.
Request DemoOverall Sentiment
strongly positive
Sentiment Score
0.72
Ticker Sentiment