Caterpillar's next growth engine: fueling AI data centers and digital infrastructure
For decades, Caterpillar meant excavators, dozers, and haul trucks. Today, the company's momentum is shifting toward AI-driven infrastructure-supplying the electrical backbone for data centers and high-density compute.
This pivot is showing up in results and deal flow, and it matters for builders, developers, and owners looking for the next cycle of work. It also explains why the stock is tracking its strongest annual run in eight years.
What changed
Caterpillar's Energy & Transportation unit reported a 25% year-over-year increase in sales to users, with energy generation for data centers up 33%, driven by demand for reciprocating engines. That surge lines up with a major engagement: equipping Joule's High Performance Compute Data Center Campus in Utah with distributed generation capable of 4 gigawatts for the Intermountain West.
The site's engine lineup centers on Caterpillar's G3520K generator sets plus support equipment. The system also captures waste heat to help cool dense server racks-important for sites targeting scale without overwhelming local utilities.
Why this matters to real estate and construction teams
Data centers are now a durable source of work across site development, MEP, civil, and mission-critical trades. The bottleneck isn't demand-it's electrical capacity, interconnect timelines, and equipment availability.
As more campuses add onsite generation and heat reuse, projects look less like simple equipment installs and more like integrated energy plants. That creates opportunity for firms that can coordinate entitlements, fuel logistics, thermal design, and fast-track delivery.
- Plan for multi-phase electrical capacity (tens to hundreds of MW) with staged generator sets and switchgear.
- Design for N+1/2N redundancy, emissions compliance, noise mitigation, and vibration control.
- Secure fuel strategy early (natural gas lines or diesel storage), and align with air permits.
- Evaluate heat recovery to cut cooling load and lower operating costs.
- Lock utility coordination and interconnect queue strategy before land closes.
- Sequence construction so generation, cooling, and IT halls come online in waves to pull revenue forward.
Manufacturing expansion that supports delivery
To meet demand, Caterpillar announced a $725 million expansion of its large-engine facility in Lafayette, Indiana-one of the biggest single manufacturing investments in company history. The goal: materially increase engine output for customers who need reliable electricity during outages and steady capacity for AI, data centers, and other critical sites.
For project teams, this is a signal. Lead times can still be long, but supply is being added where it's needed most.
Sector friction you still have to plan around
Higher borrowing costs have cooled big-ticket machinery purchases, and price pass-through has been harder for OEMs. Added tariffs on imported components from South Asia are another headwind.
Bernstein flagged 2025 as a tough year across industrial equipment with estimates down 5%-10%, and sees 2026 improving if policy and rates line up. Translation for builders: budget with contingencies, protect schedules with early procurement, and keep alternates ready.
Sentiment and stock context
Retail sentiment on Stocktwits slipped to "bearish" for both Caterpillar and Deere, with message volumes low. Yet shares of Caterpillar are up nearly 60% year to date, while Deere has gained about 11%.
Sentiment can lag real project pipelines. The structural need for compute doesn't care about headlines; it cares about sites, electrons, and uptime.
Practical next steps for developers, GCs, and owners
- Engage generator vendors early for specs, emissions profiles, and delivery windows. For reference on engine families used in these projects, see Caterpillar's G3520 series product page.
- Target regions with favorable siting and growing transmission access (the Intermountain West is a clear example), and bake water and thermal strategies into site selection.
- Use phased commissioning to bring revenue online sooner while major interconnects are pending.
- Upskill project managers and precon teams on AI/data center requirements to reduce redesign cycles and RFIs. If you need a quick path, browse AI courses by job.
Bottom line
Caterpillar's shift into data center energy isn't a side bet-it's a new demand engine tied to AI's climb. For construction and real estate teams, the work goes to those who can deliver electrical capacity, thermal efficiency, and speed without risking uptime.
Get in early, lock equipment, and design like an energy plant that happens to host servers.
Your membership also unlocks:
