SoftBank plans 75 billion euro AI data center buildout in France
SoftBank is committing up to 75 billion euros to build AI data centers across northern France, with 45 billion euros in facilities operational by 2031. This represents Europe's largest AI infrastructure play from the Japanese conglomerate and signals intensifying competition for compute capacity as foundation model training and inference demands accelerate. The scale underscores how infrastructure investment has become a primary battleground for AI dominance, though SoftBank's track record of announced megaprojects that fail to materialize warrants skepticism on full execution.
Modelwire context
Analyst takeThe France-specific geography matters more than the headline number: northern France offers proximity to undersea cable landings, relatively cheap nuclear-backed electricity, and a government actively courting hyperscale commitments as part of its industrial AI strategy. SoftBank is not just buying compute capacity, it is buying political goodwill and grid access in a jurisdiction that has been explicitly trying to pull investment away from the UK and Germany.
This is largely disconnected from recent activity in our archive, as we have no prior coverage of SoftBank infrastructure moves or European data center policy to anchor against. More broadly, this story belongs to the emerging competition among sovereign and quasi-sovereign actors to lock in physical AI infrastructure before power and land constraints tighten further. The relevant comparison set is the wave of US hyperscaler commitments to European capacity announced through 2024 and 2025, which established the baseline SoftBank is now trying to outbid.
Watch whether SoftBank closes binding land and grid agreements in northern France within the next 12 months. Announced spend without signed infrastructure contracts is the pattern that has defined the conglomerate's prior megaproject announcements, and concrete site permits would be the first real signal this commitment is different.
This analysis is generated by Modelwire’s editorial layer from our archive and the summary above. It is not a substitute for the original reporting. How we write it.
Modelwire Editorial
This synthesis and analysis was prepared by the Modelwire editorial team. We use advanced language models to read, ground, and connect the day’s most significant AI developments, providing original strategic context that helps practitioners and leaders stay ahead of the frontier.
Modelwire summarizes, we don’t republish. The full content lives on the-decoder.com. If you’re a publisher and want a different summarization policy for your work, see our takedown page.