Why Swedish Companies Are Reconsidering Their Cloud Strategies

Security, costs and artificial intelligence needs are prompting Swedish companies to reassess where their data should be stored, according to Regina Donato Dahlström, head of Equinix in the Nordic region.

She says the discussion is no longer simply about public cloud or on-premises systems. Companies must decide which data belongs where, how systems connect and what changing requirements cost. Hybrid environments can combine private and public cloud without excluding major cloud providers, but organisations need a clear view of their data, dependencies and responsibilities.

Data-transfer charges are an important part of that calculation. AWS notes in its architecture guidance that fees can depend on traffic origin, destination and volume. Storage and computing prices alone do not show the full cost, particularly when transfers provide functions such as redundancy.

FinOps Foundations’ State of FinOps 2026 survey identifies workload optimisation and reduced resource waste as the highest-priority individual task. Economic management of AI is the leading agenda item for the future, making it important to test costs against different levels of AI use.

Donato Dahlström expects more companies to use business-critical data in AI applications, particularly in finance, biotechnology and defence. Merck presented a supercomputer in an Equinix data centre in München in November 2025. The system combines private and public cloud to adjust computing capacity and support pharmaceutical research and new semiconductor materials.

• AI infrastructure needs differ between model training and inference.
• RAG applications can retrieve relevant information from external knowledge sources.
• Reconsidering a cloud strategy does not necessarily mean moving systems.