Fossefall Moves to Scale Nordic AI Capacity

Norwegian AI infrastructure developer Fossefall has selected Armada as a technology partner for planned AI factory deployments across Norway, Sweden and Finland. The agreement begins with an order for five Leviathan modular data centers, giving Fossefall an early hardware foundation for its wider regional expansion. Those initial units will provide more than 9 megawatts of capacity and are scheduled for delivery by the first quarter of 2027. The deal carries a contract value of several hundred million Norwegian kroner.

The partnership forms part of Fossefall’s ambition to develop more than 1 gigawatt of Nordic-owned AI capacity by 2030. That target places the company among a growing group of European infrastructure developers seeking greater control over the physical systems supporting artificial intelligence. Rather than depending entirely on global cloud platforms, Fossefall is positioning its model around infrastructure ownership, regional deployment and European operational control. The strategy reflects a wider shift in AI infrastructure from simple capacity expansion toward questions of sovereignty, supply-chain resilience and economic ownership.

Fossefall Builds an Integrated AI Infrastructure Model

Fossefall plans to manage more of the infrastructure stack than a typical data center developer. Its model includes energized land, construction, operations and AI services. This approach gives the company more control from development through commercial deployment. It also keeps several parts of the value chain under one strategy. The company expects renewable energy to power its planned facilities. Its broader model combines energy resources with AI computing capacity. Fossefall also wants to create more local economic activity around these projects. The strategy could give the company greater control as demand grows.

“We selected Armada for its best-in-class technology, proven execution capabilities, and ability to significantly accelerate time to market,” said Øyvind Vesterdal, CEO of Fossefall. “We were also keen to work with a strategic partner manufacturing in Europe, to provide a resilient supply chain and reduce exposure to global disruptions.

By refining clean energy into high-tech AI capacity, we will ensure national control over critical technology, create a billion-kroner export industry and produce significant local economic benefits.”

European manufacturing was an important factor in the decision. AI projects often face long equipment lead times and supply-chain constraints. Developers must also coordinate power, construction and computing equipment. Consequently, Fossefall sees the partnership as both a deployment and resilience strategy.

Nordic AI Factories Keep GPUs Under European Control

The agreement covers the modular data center platform. Fossefall will separately own and operate the GPU infrastructure inside the Leviathan units. This structure gives the company direct control over its critical computing assets. It also supports its wider sovereignty strategy. The model differs from approaches that rely entirely on external cloud infrastructure. Fossefall wants to retain greater control over where AI workloads run. It also wants more of the resulting economic value to remain in Europe. The strategy connects infrastructure ownership with AI independence.

“Sovereign AI Is a foundational requirement for national competitiveness, security, and economic resilience,” said Dan Wright, Co-founder and CEO of Armada. “We’re giving Fossefall the speed and scale to execute that vision on Norwegian terms.”

Sovereign AI is becoming an infrastructure challenge as well as a policy issue. AI ambitions require access to power, GPUs and data center capacity. They also require technical skills and reliable operations. Fossefall wants to bring those elements together under one regional model. The company also plans to involve local firms in site preparation and installation. Local companies could also support facility operations. Armada will help train and support employees for specialized deployment work. This approach could create a stronger local operating ecosystem.

Modular Design Could Shorten the AI Factory Build Cycle

The Leviathan systems come from Armada’s Galleon family of modular data centers. Armada manufactures the systems in Dublin, Ireland. The facilities follow NVIDIA reference architecture. They are also designed for different environmental conditions. Fossefall selected the modular format to accelerate capacity deployment. The company can adapt the systems to local project requirements. Armada says the units can be delivered in months rather than years. That speed could become important as AI demand continues to rise. Traditional data center projects often require extensive civil works. Developers must also secure permits and complete major power infrastructure. Those steps can delay the arrival of computing capacity. Modular infrastructure offers a different deployment path.

Moreover, modular deployment can support phased infrastructure growth. Companies can add capacity in smaller and repeatable increments. This model can better align spending with demand. It may also reduce the need for large upfront investments. The Leviathan design also aims to simplify operations. Armada says its closed-loop cooling system eliminates daily water consumption. Localized electrical routing can also simplify construction. Additional power infrastructure can be installed as new units come online.

This staged model could support Fossefall’s long-term expansion plans. The company can add infrastructure as projects develop. It may also manage capital spending more gradually. The approach will now face its first major test through the Nordic deployment.

A Strategic Test for Nordic AI Infrastructure

The first five units represent an early step toward Fossefall’s 2030 target. However, the deployment could establish a blueprint for future projects. Successful installations may allow the company to repeat the model across the region. Standardized infrastructure could support faster expansion. Fossefall plans to combine that standardized platform with local construction and operations. The model offers an alternative to building only massive centralized campuses. Instead, capacity can expand across multiple locations. That flexibility may become valuable as AI demand spreads geographically.

The project also raises a broader question for the Nordic region. Can abundant clean energy translate into locally controlled AI capacity? Fossefall is betting that infrastructure ownership will be as important as power availability. The company wants to combine both advantages. Armada provides the modular infrastructure designed to speed up deployment. Fossefall brings the regional development and ownership strategy. Together, the companies will test whether this model can scale. The key challenge is turning an initial 9-megawatt-plus deployment into a regional AI network by 2030.