Finland's €5M Bet: The Real Quantum Computing Bottleneck Is Copper Cables
Finland's €5M Bet: The Real Quantum Computing Bottleneck Is Copper Cables
If a million-qubit quantum computer ever runs, it will need power comparable to a nuclear plant — and the culprit isn't the qubits. It's the copper wires controlling them. Finland is betting €5 million that light, not electricity, is the answer.
Walk into any quantum computing lab today and you'll find the same unglamorous bottleneck: thousands of coaxial cables snaking from room-temperature electronics into a chandelier-like dilution refrigerator, cooling superconducting qubits to fractions of a degree above absolute zero. Those cables carry the microwave signals that tell each qubit what to do. They also carry heat. At millikelvin temperatures, that heat load is catastrophic.
"A million-qubit quantum computer could require energy comparable to the output of a nuclear power plant," says Associate Professor Jukka Viheriälä of Tampere University, who leads the Finnish consortium project QScale. "Due to the thermal load generated by quantum computers, increasing their performance is not currently feasible."
That is the wall the quantum industry is approaching — and Finland is spending €5 million to break through it.
The QScale project
QScale — "Scaling up quantum computing by telecom-based technologies" — launches in September 2026 with three-year funding from Business Finland's Rise to the Challenge programme. The project is coordinated by VTT Technical Research Centre of Finland, with Tampere University and Aalto University as consortium partners. Tampere's share is approximately €1.6 million.
The goal is to replace the electrical cabling that currently controls qubits with ultrafast optical data transmission. Instead of routing microwave signals down coaxial cables — each one a tiny radiator generating noise in the ultra-cold environment — QScale aims to send control signals via fiber optics, converting them to microwave form only at the point of need, inside the refrigerator.
"If we can overcome this bottleneck, quantum computing can genuinely transition from the experimental stage to industrial-scale applications and enable breakthroughs in diverse fields, such as medicine, materials science, optimisation and artificial intelligence," Viheriälä says.
The thermal math nobody wants to do
Today's largest quantum processors — IBM's 1,000+ qubit Condor, Google's Willow, the systems from IQM and others — are already running into power and thermal constraints that have nothing to do with the qubits themselves. The control electronics outside the refrigerator generate kilowatts of heat. The dilution refrigerator can only handle a fraction of that. Scale to a million qubits, and the numbers become absurd.
The QScale proposal, grounded in published modeling, puts the million-qubit energy demand in nuclear-plant territory. The bottleneck isn't qubit coherence or gate fidelity — it's heat leaking in through the wiring. Every coaxial cable carrying a control signal is also carrying thermal noise into the coldest place in the universe.
Optical fibers carry data with orders of magnitude less heat load per channel. They also enable higher-density parallelization, since fiber bundles are thinner and more flexible than equivalent coaxial assemblies. The physics has been understood for years. The engineering challenge — making it work inside a dilution refrigerator at the precision required for quantum operations — is what nobody had solved.
The proof of concept already exists — in Helsinki
Six months before QScale launches, QphoX and Bluefors demonstrated optical control of a superconducting transmon qubit at the Bluefors Quantum Applications Lab in Helsinki. The result, posted to arXiv (2603.18780) and announced March 24, 2026, is the first known demonstration of an optical control system operating inside a dilution refrigerator with a real qubit — not just at the component level, but in an integrated, "out-of-the-box" configuration.
The QphoX optical control system achieved compatibility with the Bluefors cryogenic platform without degrading qubit performance. "The results show the potential of the optical qubit control technique to break through current cryogenic thermal bottlenecks without compromising on performance," the companies said.
That matters enormously for QScale. The project is not proposing something unproven. It is building on a first demonstration to tackle the engineering scale-up: higher-speed optical-to-microwave conversion at millikelvin temperatures, cryogenic photonic packaging, and system-level stability for high-fidelity, repeatable qubit operations.
What QScale will actually build
Within QScale, Tampere University will develop high-frequency microwave signal generation and advanced quantum packaging technologies. The project draws on Tampere's existing semiconductor and chip infrastructure — the SiPFAB pilot line, the ChipIn characterisation facility, and the SoC Hub — for subsequent piloting and industrial deployment. Aalto University brings quantum hardware expertise, including its Q20 quantum computer operated jointly with IQM.
The first phase runs September 2026 through August 2029. Additional funding may follow for a two-year extension phase focused on industrial partner integration.
The ASML question
The most important second-order consequence of solving cryogenic photonic interconnects isn't the energy bill for quantum computers. It's the supply chain position.
ASML became the most critical — and most politically contested — company in semiconductor manufacturing by mastering extreme ultraviolet lithography, a domain so difficult that no other company on Earth could replicate it. Whoever builds reliable, manufacturable cryogenic photonic control systems for quantum computing occupies a structurally similar position. Every superconducting quantum computer that needs to scale past the current heat wall will need it.
QphoX, based in Delft, Netherlands, is currently the closest company to filling that role. Bluefors, the Helsinki cryogenic systems company, is the other half of the emerging Nordic photonic quantum control stack. QScale is Finland's attempt to own part of the underlying research and, eventually, the intellectual property.
Finland's quantum sovereign bet
The US-China quantum race is typically framed as a bipolar contest: American companies on one side, Chinese state labs on the other. Finland's QScale reflects a different strategic logic — a small, highly specialized country identifying the infrastructure chokepoint both sides will eventually need and trying to own it.
Finland ranked second globally for quantum readiness in ECIPE's 2024 assessment, behind only the United States and ahead of every EU member state. That ranking reflects decades of sustained investment: VTT's quantum computing programme, the Aalto-IQM Q20 machine, the Finnish quantum ecosystem anchored in Helsinki and Tampere. The €50 million SUPREME project — 23 partners across 8 EU states, coordinated by VTT, targeting a 200-qubit 3D-integrated module — runs in parallel, addressing the same scaling problem from the qubit architecture side.
QScale addresses it from the infrastructure side. Together, they give Finland a position in the million-qubit era that doesn't depend on licensing American or Chinese control electronics.
The real test
The physicists will tell you the qubit is the hard part. The engineers will tell you it's everything else — the wiring, the packaging, the thermal management, the control software — that determines whether a quantum computer stays in the lab or changes the world. QScale is a €5 million bet that the latter problem is, in some sense, the more tractable one. And that solving it is worth building a country's quantum strategy around.
The project launches in September. The stakes are measured in megawatts.
Primary sources: Tampere University press release (June 1, 2026) · QphoX/Bluefors announcement (March 24, 2026) · VTT SUPREME announcement (May 11, 2026)