The Quiet War Over Quantum's Coldest Real Estate

In 1911, a Dutch physicist named Heike Kamerlingh Onnes cooled mercury to minus 269 degrees Celsius and watched its electrical resistance vanish entirely. He had discovered superconductivity. It took another century for that discovery to become the foundation of the most expensive appliances on Earth: dilution refrigerators, the massive cryogenic systems that keep quantum processors cold enough to function.

Today, every serious superconducting quantum computer runs inside one of these machines. They cost between $500,000 and $2 million each. They stand six feet tall. They look like inverted golden chandeliers wrapped in wiring and thermal shields. And they must hold their qubits at roughly 15 millikelvin (colder than outer space by a factor of roughly 180). If the temperature drifts even slightly, the quantum state collapses and the computation is garbage.

This is the bottleneck nobody talks about. The quantum computing conversation in 2026 orbits around qubit counts, error correction breakthroughs, and who will reach fault tolerance first. But underneath all of it sits a mundane engineering constraint: can you actually keep the hardware cold enough, stable enough, and connected enough to run real workloads? The fridge problem isn't glamorous. It doesn't make headlines. But it gates everything.

Here's the current state of play. Google's Willow chip, unveiled in late 2024, demonstrated below-threshold error correction on a 105-qubit system. IBM's Heron processors are being wired together through their modular architecture. Rigetti Computing (NASDAQ:RGTI) continues to iterate on its own superconducting designs. Each of these efforts requires cryogenic infrastructure that is shockingly hard to scale. You can't just buy a bigger fridge. The wiring alone (hundreds of coaxial cables running from room temperature down to 15 millikelvin) introduces heat at every stage. Every additional qubit means more control lines, more heat load, and more thermal management headaches.

The companies building quantum processors know this. They've started vertically integrating, designing custom cryogenic components, building their own signal delivery systems, even exploring alternative qubit modalities that operate at warmer temperatures. Trapped-ion systems from Quantinuum (a Honeywell spinoff) run at room temperature in principle, though they bring their own scaling challenges. Photonic approaches sidestep the cold problem entirely but struggle with deterministic gate operations.

Meanwhile, the market has been doing something odd. After the euphoric run in late 2024 and early 2025 (when quantum stocks tripled and quadrupled on the back of Google's Willow announcement), the sector corrected hard. Most pure-play quantum names gave back 40 to 60 percent from their peaks. Then in early 2026, a second wave of buying hit, driven by government procurement signals and the Pentagon's growing anxiety about quantum-vulnerable encryption. The National Institute of Standards and Technology finalized its post-quantum cryptography standards, and suddenly every defense contractor, bank, and cloud provider needed a quantum strategy.

But the money flowing into the sector is splitting into two very different buckets. Bucket one: hardware companies trying to build the actual quantum computer. These are moon shots. The timelines are long, the cash burn is real, and the revenue is mostly government grants and small pilot contracts. Bucket two: companies building the tools, components, and security layers that the entire quantum industry needs regardless of which hardware approach wins. Think picks and shovels during a gold rush, except the picks are cryogenic control systems, quantum-safe encryption, and chip testing equipment.

One company in the second bucket has been doing something unusual. It doesn't build quantum processors. It doesn't promise to solve protein folding or break RSA encryption. Instead, it builds the photonic networking technology that quantum systems will need to communicate with each other and with classical infrastructure. Its founding thesis was simple but prescient: quantum computers will eventually need to talk to each other, and when they do, the interconnect layer will be worth more than the processors themselves. The company's technology enables quantum key distribution (QKD), a method of encrypting data using the laws of physics rather than mathematical assumptions. It has signed contracts with NATO members. It has partnered with European telecom operators. And it remains, by most measures, badly misunderstood by a market fixated on qubit counts.