SoftBank Commits €75 Billion to Build 5 GW of AI Data Center Capacity in France

SoftBank Group announced a commitment to develop and operate 5 GW of AI data center capacity in France with an investment reaching €75 billion ($87 billion), marking the Japanese conglomerate's largest infrastructure commitment in Europe. The announcement came during the 2026 Choose France summit hosted by President Emmanuel Macron.

The first phase comprises an initial €45 billion investment to deliver 3.1 GW of AI data center capacity in the Hauts-de-France region by 2031. The deployment will span three locations: Dunkirk (Loon-Plage), Bosquel, and Bouchain, with facilities designed specifically for AI inference and training workloads.

Infrastructure Scale and Geographic Distribution

The 5 GW total capacity represents substantial computing infrastructure by contemporary standards. For context, individual hyperscale data centers typically operate between 100-300 MW, making SoftBank's planned deployment equivalent to roughly 15-50 conventional facilities, though AI-optimized infrastructure generally requires higher power density configurations.

SoftBank will partner with Sesterce to develop a 1 GW facility in Bosquel as part of the broader regional buildout. The Hauts-de-France region's selection likely reflects its proximity to submarine cable landing points, existing power grid infrastructure, and favorable regulatory environment for large-scale industrial projects.

The Dunkirk site will serve dual purposes, housing data center operations while anchoring advanced data center manufacturing capabilities in partnership with Schneider Electric. This manufacturing component suggests SoftBank intends to localize supply chain elements rather than importing all infrastructure components.

Investment Timeline and Financing Structure

The €75 billion commitment spans multiple phases, with the initial €45 billion targeting the first 3.1 GW by 2031. The remaining 1.9 GW of capacity presumably follows in subsequent phases, though SoftBank has not disclosed specific timelines for the complete 5 GW deployment.

This investment scale approaches SoftBank's entire current market capitalization and significantly exceeds the group's typical single-project commitments. The financing structure remains unclear, though SoftBank CEO Masayoshi Son has historically leveraged debt markets and asset sales to fund Vision Fund investments and major infrastructure projects.

Reports suggest Son initially floated investment figures as high as $100 billion during discussions with French officials, indicating the final €75 billion represents a negotiated commitment rather than an opening bid.

Strategic Context and Market Positioning

The France deployment marks SoftBank's first major European data center commitment, expanding beyond its traditional geographic focus in the United States, Japan, and broader Asia. Son specifically cited President Macron's personal commitment to ensuring France's economic success as influential in the decision, despite SoftBank's historical investment concentration elsewhere.

"AI is entering a new era, and the countries that build the infrastructure for this transformation will shape the future of technology, industry and society," Son stated regarding the investment.

This framing aligns with SoftBank's broader AI infrastructure thesis. The group has positioned itself as a primary financier of AI development through its Vision Fund vehicles and direct investments in companies like Arm Holdings, which designs processors used extensively in AI training and inference.

Looking at what this means for the broader market, the commitment signals institutional confidence in sustained AI computing demand beyond current hypergrowth phases. Five gigawatts of capacity suggests SoftBank anticipates AI workloads will continue expanding at rates that justify massive infrastructure pre-positioning, rather than incremental capacity additions matching near-term demand.

We have seen this pattern before, when cloud providers like Amazon and Microsoft built data center capacity years ahead of enterprise migration timelines, betting correctly that application modernization would eventually fill available infrastructure. SoftBank appears to be making a similar forward bet on AI adoption curves.

Technical and Regulatory Implications

The AI-specific designation suggests these facilities will differ from traditional cloud data centers in power distribution, cooling systems, and network architecture. AI training workloads typically require high-bandwidth interconnects between compute nodes, while inference workloads prioritize low-latency connectivity to end users.

France's regulatory environment offers certain advantages for large-scale AI infrastructure deployment, including EU data residency requirements that encourage local processing and the country's substantial nuclear power generation capacity, which provides relatively stable electricity pricing for power-intensive operations.

The partnership with Schneider Electric for manufacturing capabilities also positions the project to benefit from EU industrial policy initiatives promoting local technology production, potentially including subsidies or tax advantages for domestically manufactured infrastructure components.

Market Impact and Competitive Response

The scale of SoftBank's commitment likely pressures other major infrastructure investors to accelerate their own European AI data center plans. Microsoft, Google, and Amazon have announced various European capacity expansions, but none approach the 5 GW scale SoftBank has committed to in a single country.

The investment also represents a significant validation of France's efforts to position itself as a European AI hub, competing with established technology centers in Ireland, the Netherlands, and Nordic countries for hyperscale infrastructure deployments.

For the AI industry more broadly, the commitment provides additional evidence that infrastructure bottlenecks may ease as private capital flows into purpose-built facilities. Current AI development faces constraints from limited access to specialized computing resources, particularly for training large language models and other foundation model architectures.

The broader context here suggests SoftBank views AI infrastructure as a multi-decade investment thesis rather than a near-term opportunity. The 2031 timeline for initial capacity delivery and the massive financial commitment indicate Son and SoftBank believe current AI adoption represents an early phase of a longer transformation cycle, rather than a peak demanding immediate capacity response.