Europe's Grid Nerve Center Is Getting a Private

Nobody asked, in any public forum I can find, whether a single private company should own and operate the grid stabilization layer for a continent-spanning electrical network. That decision, if it was made at all, was made in a boardroom — and the regulatory architecture that would ordinarily apply to something this consequential appears to have arrived late, or not at all.

That's the question Flexbase forces into the open. And it deserves a sharper frame than "ambitious project" or "engineering marvel," both of which it undeniably is.

What Is Actually Being Built

Start with the engineering, because it earns the policy anxiety that follows.

The project, backed by developer Marcel Aumer, proposes to construct the world's largest vanadium flow battery beneath the town of Laufenburg in the Swiss canton of Aargau — nine stories underground, covering more than 215,000 square feet of excavated footprint, targeting an output of 1.2 gigawatts. As the Engineering The Impossible documentary on the project puts it: "This single buried machine is built to push out as much power as a full-size nuclear reactor."

That is not a metaphor deployed for effect. That is the engineering specification.

Vanadium flow batteries store energy not in solid cells but in liquid electrolyte — vanadium dissolved in a water-based solution — pumped between tanks as the battery charges and discharges. The chemistry is nonflammable, which matters enormously when you're discussing a structure buried beneath an inhabited town. Unlike lithium-ion installations, which have a history of thermal runaway events that have complicated permitting worldwide, vanadium flow systems don't catch fire. That property, more than anything else, is why Aumer chose this chemistry.

The engineering problems created by burying this thing are genuinely extraordinary. The pit sits on the southern bank of the Rhine, where the ground is loose valley gravel — porous and waterlogged. Engineers had to construct a concrete containment system before excavation could begin, essentially building a tub in the ground to hold the river out. Groundwater pressure underneath an empty concrete shell creates a second problem: without countermeasures, the entire foundation would behave like a cork in a bucket and float upward. The solution involves massive steel anchor rods drilled deep through the gravel and locked beneath a thick concrete slab — tent stakes, as the documentary calls them, holding down a structure that will eventually weigh as much as a cruise ship.

There is also the matter of the liquid itself. Because the energy is stored in fluid, and because that fluid shifts from tank to tank as the battery cycles, the center of gravity of the entire structure moves continuously. Almost no building in human history has been engineered to account for tens of thousands of tons of mass in constant lateral motion. The engineers' answer is to overbuild everything — thicker slabs, more steel than the calculations strictly require — because the math for a building that sloshes doesn't have clean precedent.

Construction sequence adds another wrinkle: the storage tanks are too large to fit through any door, so they go in before any roof exists, lowered by crane onto the open foundation, with each successive floor poured over them — a layer-cake assembly that works downward through the earth rather than upward through the sky.

Completion is targeted for 2028. Grid connection is expected around 2029. The documentary is candid that industry observers are watching those dates with skepticism proportional to the project's ambition.

The Demand Context

The grid stress that makes a project like this commercially legible is real and documented. According to the IEA's Electricity 2026 report, data center electricity demand is projected to approach 1,000 terawatt-hours globally by the end of the decade. For context, Statista data on UK household consumption puts 2022 global data center demand — which ran between 240 and 340 terawatt-hours — at roughly the equivalent of the United Kingdom's entire annual electricity use. The IEA's projection represents roughly a tripling of that baseline.

The grid was designed for predictable, cyclical demand. AI inference workloads don't have the behavioral regularity of factory shift patterns or residential evening peaks. The variability is the problem, and large-scale storage is the proposed solution. China demonstrated the concept's viability at scale: according to Interesting Engineering, a gigawatt-hour-scale vanadium flow battery system came online in China at the end of 2024, at that point the largest of its kind ever built. Switzerland's Flexbase project is designed to exceed it — and to do so at a location that carries specific strategic weight for the European grid.

Laufenburg is, by any historical account, a foundational node in Europe's interconnected electrical network. The documentary describes it as the birthplace of the continental grid, where the power systems of multiple nations were first wired together — and notes that Laufenburg has retained its function as an anchor point ever since. It sits on the same stretch of the Rhine as the Leibstadt nuclear plant. When Flexbase describes this as "the beating heart of an entire continent's electrical system," the characterization is not marketing copy.

The Governance Gap

Which brings me to the question nobody in this project's promotional materials seems eager to answer.

The EU's critical energy infrastructure framework — operationalized partly through the NIS2 Directive, which mandates cybersecurity standards across essential services including energy, and guided operationally by ENTSO-E, the European network of transmission system operators — is built around the premise that critical nodes require oversight commensurate with their systemic importance. ENTSO-E coordinates grid planning and security across the continent and publishes adequacy assessments that grid operators rely on for capacity planning.

Flexbase is privately financed. Cost estimates in the documentary range from $1.2 billion to north of $6 billion — a spread so wide it should raise eyebrows in any due diligence review, let alone a publication covering energy infrastructure. The project appears to carry no public-sector backstop. If construction stalls in 2027 because the economics deteriorate or the geotechnics turn hostile, what is the liability structure? Who holds the contractual obligation to the grid operators who are presumably incorporating this 1.2-gigawatt capacity into their planning scenarios? What happens to the hole?

These are not hypothetical concerns dressed up as rigor. They are the precise questions that EU critical infrastructure governance is supposed to address — and the answers, if they exist, are not publicly available.

NIS2, which EU member states were required to transpose into national law by October 2024, covers operators of essential services in the energy sector, including electricity generation and distribution. Whether Flexbase falls within NIS2's scope — and under which national transposition, given Switzerland's complicated relationship with EU law — is a regulatory question with real consequences. Switzerland is not an EU member, which means the EU's critical infrastructure directives apply at the border rather than the installation. Swiss federal energy law governs domestically. But the grid connection Flexbase is designed to serve does not stop at the Swiss border, and ENTSO-E's system adequacy models presumably don't either.

The documentary notes, approvingly, that Laufenburg's strategic importance is precisely why burying the battery makes sense: "If you're placing the single most important shock absorber on the European grid right at the most important nerve center on that grid, you don't leave it sitting out in the open exposed to floods, fire, or a single bad day." Fair enough. But physical hardening is not the same as institutional resilience. A concrete shell nine stories underground does not answer the question of what happens when the private entity responsible for it encounters financial, legal, or operational distress.

China's comparable installations sit above ground, in open desert, documented and photographed. Switzerland's approach — go bigger, go underground, go private — is a coherent engineering choice. Whether it is a coherent governance choice is a separate matter entirely, and one that the project's advocates have been conspicuously quiet about.

The hole in Laufenburg is undeniably real. The regulatory framework capable of governing what goes inside it is considerably harder to locate.

Samira Barnes is a tech policy and regulation correspondent for Buzzrag.