---
title: The Regenerative ‘AI Factory’ Model And The Push For On-Site Renewables
description: AI data centres move to rural Montana with renewable energy, liquid cooling and heat reuse–reshaping industry trends and boosting sustainable development
author: Darie Nani (Editor-in-Chief)
date: 2025-09-10T15:50:29.000Z
updated: 2026-02-26T18:02:02.852Z
canonical: https://www.sovereignmagazine.com/article/the-regenerative-ai-factory-model-and-the-push-for-on-site-renewables
image: https://cdn.nanimediahouse.com/vi1hxpw6hyw.jpg
categories: Artificial Intelligence
content_type: Feature
region: Montana
publication: Sovereign Magazine
---

[AI workloads are pushing data centres](https://www.sovereignmagazine.com/article/ai-energy-use-to-more-than-double-electricity-demand-by-2026) to consume unprecedented amounts of power, yet most facilities remain clustered in urban areas, tethered to traditional grid infrastructure and disconnected from the communities around them. As electricity companies plot [massive 142% bill increases](https://www.dailymail.co.uk/yourmoney/consumer/article-14893697/Electricity-companies-plot-massive-142-bill-hike-despite-increasing-risk-blackouts.html) to pass on the cost of energy-hungry data centres, operators are being forced to reconsider the fundamentals of where and how they build. KRAMBU’s new AI facility in Montana offers a different template: rural sites powered by on-site renewables and designed to give back to their surroundings.

## The Montana Project – What’s New?

KRAMBU’s Montana campus combines direct liquid cooling with renewable energy resources including solar, biomass-to-energy and green hydrogen generation. The facility eliminates traditional refrigerants entirely, instead using direct liquid cooling systems that circulate coolant directly to heat sources. What makes this approach particularly interesting is the heat recovery component – waste heat from the servers gets captured and repurposed for agricultural, industrial and commercial use across the broader property.

The site operates under what KRAMBU calls an ‘Industrial Symbiosis Model’, where the data centre becomes part of a larger system rather than an isolated facility. Plans include additional compute halls, co-located industries and expanded on-site energy generation to reduce dependence on traditional grid infrastructure.

## What It Means For Site Location And Rural Areas

Moving AI infrastructure to rural Montana addresses several problems that urban data centres face. [Limited land and power availability](https://www.cbre.com/insights/reports/global-data-center-trends-2024) in urban areas increase costs and delay developments, while rural locations offer better power availability and opportunities for renewable energy integration. Montana specifically provides tax incentives including property tax breaks and no sales tax for data centre development.

The state’s approach to repurposing stranded industrial sites creates opportunities that wouldn’t exist in developed urban areas. These former industrial locations already have some infrastructure in place and face less competition for power and land resources. [Several contaminated industrial sites](https://www.epa.gov/superfund-redevelopment/superfund-sites-reuse-montana) in Montana are being repurposed for commercial and light industrial uses, with environmental monitoring ensuring safety standards are met.

KRAMBU’s heat recovery approach builds on successful examples elsewhere. [Data centre waste heat](https://intelligence.uptimeinstitute.com/resource/heat-reuse-management-primer) at 25-40°C can be used for greenhouses, hydroponic farms and aquaculture, with heat pumps intensifying the temperature when needed. Norway’s land-based lobster farming operations already use data centre heat, and there’s a projected $407 billion future market for farmed fish and seafood that could benefit from these heat reuse strategies.

‘This project represents more than a data center—it’s a regenerative system,’ said Travis Jank, President and Founder of KRAMBU. The concept turns traditional data centres into integrated facilities that support local agriculture and industry rather than simply consuming resources. [AI factories](https://www.sovereignmagazine.com/article/ai-factories-are-the-new-data-centres) like this represent a fundamental shift in how computing infrastructure serves communities.

Rural North America particularly suits this model because agricultural and industrial heat users are more readily available compared to urban district heating systems. The approach reduces reliance on fossil fuels for heating while creating revenue streams that help offset operational costs.

## Engineering Details And Business Relevance

Direct liquid cooling delivers substantial operational advantages over traditional air cooling. The technology transfers heat [50 to 1,000 times more effectively](https://www.applieddigital.com/insights/why-modern-ai-data-centers-are-harnessing-the-power-of-liquid-cooling) than air cooling, reducing power consumption by about 15% and cooling energy use by up to 40%. [Advanced cooling systems](https://www.sovereignmagazine.com/article/why-is-free-cooling-good-for-your-data-centre) like these are becoming essential for modern data centres. NVIDIA’s liquid-cooled Blackwell platform delivers up to 30x energy efficiency compared to traditional air-cooled systems.

The business case becomes compelling when considering that [liquid-cooled systems can offer up to 25x cost savings](https://blogs.nvidia.com/blog/blackwell-platform-water-efficiency-liquid-cooling-data-centers-ai-factories/) despite higher initial infrastructure investment. The technology also supports higher rack densities and extends hardware lifespan, critical factors for AI workloads that demand intensive computational power.

On-site renewable generation addresses grid reliability concerns while reducing operational expenses. Companies like Soluna and IREN already co-locate data centres with curtailed renewable energy in rural Kentucky and Texas, capitalising on excess renewable power that would otherwise go unused. [Renewable energy integration](https://www.sovereignmagazine.com/article/can-ai-and-renewables-make-crypto-mining-pay-business-prospects-and-pitfalls-of-a-new-platfor) is becoming increasingly important for energy-intensive computing operations.

‘With liquid cooling, on-site renewables, and integrated community benefits, we’re building infrastructure that works in harmony with its environment—and becomes a long-term asset for the region,’ said Steve Wood, CEO of KRAMBU.

## Possible Trade-Offs Or Challenges

Rural data centre development faces distinct obstacles. [Zoning regulations can cause approval difficulties](https://www.datacenterknowledge.com/regulations/land-barriers-how-zoning-regulations-could-stall-data-center-industry-expansion) and the majority of Montana lacks easy access to large fibre optic networks. Environmental and community concerns about energy consumption, noise and land use remain significant considerations.

The question remains whether [on-site renewables and liquid cooling](https://www.sovereignmagazine.com/article/what-happens-when-teachers-not-tech-set-the-pace-for-ai-in-classrooms) scale as easily as urban, grid-connected sites. While rural locations offer space and cooling climate advantages, they require substantial upfront investment in both renewable energy infrastructure and skilled workforce development. Slow permitting processes, particularly for technologies like carbon sequestration and hydrogen generation, could delay implementation.

Heat reuse strategies work well for specific applications but may not provide consistent revenue streams year-round. Agricultural heat demand varies seasonally, and finding suitable industrial partners requires careful planning and long-term agreements.

## Where Does This Model Fit In?

KRAMBU’s approach aligns with broader industry trends toward decentralisation. [Data centre development is shifting](https://www.bdcnetwork.com/home/news/55234915/10-biggest-impacts-to-the-data-center-market-in-20242025) from urban to rural locations driven by lower energy costs and colder climates, while over 40% of larger enterprises are expected to integrate edge computing by 2025. [Server consolidation](https://www.sovereignmagazine.com/article/server-consolidation-drives-major-energy-savings-in-modern-data-centres) and improved efficiency are driving many of these changes.

This decentralisation supports the rise of [edge computing](https://www.sovereignmagazine.com/article/the-smart-building-boom-how-openai-is-reshaping-the-automation-industries), which relocates computational power closer to data sources. As AI workloads continue growing and [Google invests $25 billion](https://www.nbcnewyork.com/news/business/money-report/google-to-invest-25-billion-in-data-centers-and-ai-infrastructure-across-largest-u-s-electric-grid/6335811/) in data centres and AI infrastructure, the pressure to find sustainable, cost-effective solutions will only increase.

The success of projects like KRAMBU’s Montana facility will likely influence how other operators approach future builds, especially as energy costs rise and environmental scrutiny intensifies. Whether this model can scale beyond individual projects to become an industry standard will depend on proving both the technical feasibility and economic viability of rural, renewable-powered AI infrastructure. [midsize operations in the energy transition](https://www.sovereignmagazine.com/article/midsize-wind-power-how-freen-s-new-turbine-targets-local-business-needs-in-estonia-and-beyond)