How AIP and Caterpillar Fuel the AI Data Centre Power Shift

The exponential growth of AI is driving unprecedented demand for reliable, high-capacity power infrastructure.

As hyperscale data centres struggle to secure adequate energy from public grids, dedicated generation platforms are becoming essential.

American Intelligence & Power Corporation has formed a strategic alliance with Caterpillar and Boyd CAT to deliver large-scale dedicated power for hyperscale AI data centres.

The agreement supports the Monarch Compute Campus in West Virginia, a multi-phase platform designed to meet growing AI-driven data centre energy demands.

Under the purchase agreement, American Intelligence & Power Corp has ordered 2GW of fast-response natural gas generator sets from Caterpillar.

Equipment deliveries are scheduled between September 2026 and August 2027, with power delivery beginning in 2026 and 2GW expected online in 2027.

The Monarch Compute Campus is positioned as a behind-the-meter solution for hyperscale operators seeking long-term access to high-reliability power without increasing pressure on public grids.

Fast-response generation technology

AI workloads present unique infrastructure challenges.

Extreme load variability and high-density compute environments require fast-response power delivery capable of adapting to rapid demand fluctuations.

The Monarch platform combines natural gas generation with battery energy storage systems designed to absorb rapid load swings while maintaining power quality.

This hybrid approach represents a template for how critical digital infrastructure adapts to the computational intensity of machine learning and large language model training.

"Our design is purpose-built for AI data centre operations, combining fast-response natural gas generation with battery energy storage to manage rapid load variability and deliver consistent power quality at scale," says Daniel J. Shapiro, CEO of American Intelligence & Power Corp.

The power platform centres on Caterpillar G3516 fast-response natural gas generator sets.

These units can ramp from zero to full load in approximately seven seconds, addressing the fundamental requirement of AI workloads that fluctuate sharply.

"Projects like Monarch demonstrate how Caterpillar's natural gas generation platforms are being deployed as core infrastructure for data centres where reliability, speed of deployment and lifecycle performance are critical," says Melissa Busen, Senior Vice President of Electric Power at Caterpillar.

The generators incorporate advanced emissions controls, including selective catalytic reduction, to meet ultra-low emissions standards.

For data centre operators balancing sustainability targets with power availability, emissions performance is increasingly important.

Local deployment and support

Boyd CAT will deliver and support the Monarch deployment, providing technical expertise and long-term service capability across the equipment lifecycle.

"The scale of the Monarch project demands precision and performance, and Boyd CAT is ready to deliver both as AIP Corp brings new generation capacity online," says Andrew Boyd, President & CEO of Boyd CAT.

The alliance establishes a framework for ongoing collaboration, including planning for phased expansion, operations and maintenance strategy, and service readiness.

The initial 2GW deployment represents the first phase, with Monarch targeting up to 8GW of total generation capacity over time.

Infrastructure model for AI

Monarch is designed as a fully self-supplied power platform that does not require incremental utility transmission or distribution infrastructure.

Power is generated onsite, reducing dependency on constrained regional grids and enabling faster deployment of data centre capacity.

With its existing West Virginia microgrid designation, the project is structured to avoid increasing rates for existing utility customers while contributing to broader grid resilience.

The Monarch model reflects a growing shift toward dedicated generation platforms that align energy delivery directly with data centre development timelines.

For hyperscale operators facing delays due to power availability, this approach offers a pathway to secure reliable capacity.

As AI continues to drive computational demand, the infrastructure supporting it must evolve to match both the scale and technical characteristics of these next-generation workloads.