Quantum Computing Has An Industrialisation Problem

Quantum computing has spent years being measured by extraordinary things. More qubits. Lower error rates. Longer coherence times. Better algorithms. Each milestone gives researchers another way to show that these machines are becoming more capable. But there's another measure of progress that's much less exciting. 

Can you build another one? Not another experimental system that takes a specialist team months to assemble and tune. Another machine built to the same standard, using components that can be sourced reliably, installed somewhere else and supported after it arrives. That's a very different challenge. It's also becoming harder to ignore. 

In May 2026, the US Department of Commerce announced $2.013 billion in planned CHIPS incentives for nine companies working across quantum computing. The programme wasn't only aimed at better processors. It identified manufacturing problems across foundries, photonics, control hardware, cryogenic systems, packaging, readout electronics and systems integration. 

Since then, several of those plans have started turning into real manufacturing investments. Quantinuum finalised a $100 million CHIPS award in September, while IBM subsidiary Anderon secured up to $1 billion for a dedicated quantum wafer foundry. Quantum computing's scientific problems haven't disappeared. 

Fault tolerance, error correction and useful applications still present enormous challenges. But solving them won't automatically create a commercial industry. Quantum computing also has to learn how to become something that can be made.

Building A Quantum Computer Isn't The Same As Manufacturing One

There's a big difference between proving you can build something and proving you can manufacture it. A research team can spend enormous amounts of time getting one machine to work. Components can be adjusted individually. Problems can be solved as they appear. If something needs to be redesigned, specialists can go back and change it. 

Manufacturing asks for something harder. The process has to work again and again, with predictable quality, costs and production times. We're starting to see quantum companies think in those terms. IonQ's Superion 256 platform offers a particularly clear example. The company says it completed six chip tapeouts, essentially completed chip designs sent for fabrication, during the first half of 2026. 

Working with semiconductor foundry SkyWater, it says its design cycle fell from nine months to two and it produced 12 times more wafer lots over six months than it had at its previous foundry. Those aren't the performance numbers quantum announcements usually lead with. They're manufacturing numbers. 

Superion is also designed as a common platform for future systems, rather than another standalone machine. IonQ says prototype systems are already being built in parallel at several US facilities, with customer deliveries planned for 2027. The claims still come from IonQ, and its future production targets remain targets rather than proof of mass manufacturing. 

But the change in what the company is measuring is revealing. Quantum hardware is starting to face the same question every complex technology eventually faces: not only whether it works, but whether the process used to make it can be repeated.

The Quantum Supply Chain Is Bigger Than The Qubit

It's tempting to think quantum computing manufacturing means making quantum chips. The reality is much messier. A useful quantum computer needs an entire physical system around its quantum processor.

Depending on the technology, that can include specialised semiconductor components, lasers and photonics, control electronics, readout equipment, advanced packaging, cryogenic cooling or vacuum systems. Those pieces then have to work together. The US CHIPS programme provides an unusually useful picture of the problem because its investments target bottlenecks across that wider stack. 

The projects include work on device reproducibility, optical systems, cryogenic integration, control hardware, readout electronics, photonic loss, packaging and interconnects. The Quantum Economic Development Consortium (QED-C) describes the current quantum supply chain as custom, fragile and increasingly strategic. 

It points to dependencies on specialist technologies such as cryogenics, control electronics and photonics, with many suppliers still serving both research and commercial markets. Requirements are also changing quickly and aren't yet widely standardised. So producing more qubits won't solve the industrialisation problem by itself. 

Every surrounding component has to become reliable and available enough to support the machines those qubits eventually become part of.

There Won't Be One Quantum Factory

Even that supply-chain picture comes with a complication. There isn't one universally accepted way to build a quantum computer. Different companies are pursuing superconducting qubits, trapped ions, neutral atoms, photons and silicon spins, among other approaches. Each creates its own physical requirements. 

A trapped-ion machine, for example, depends heavily on precise optical systems. Superconducting systems have different cryogenic, material and packaging requirements. Photonic quantum computers introduce their own challenges around components such as photon detectors and low-loss optical packaging. 

The Commerce Department's funding portfolio reflects those differences. Its projects span five quantum modalities and target problems ranging from high-powered optical systems for neutral atoms to advanced dielectric materials for superconducting systems and electro-optic materials for photonic machines. 

So the emerging quantum manufacturing industry probably won't look like one factory gradually learning to produce the same product faster. Several overlapping ecosystems are developing at once. Some manufacturing capabilities may eventually be shared between them. Others will remain highly specialised. 

And they're doing this before the industry knows which quantum architectures will prove most commercially useful.

A Quantum Manufacturing Ecosystem Is Starting To Form

This is where some of the latest developments become particularly interesting. In September, Anderon finalised an agreement for up to $1 billion in CHIPS funding, alongside another $1 billion investment from parent company IBM. Its Albany facility is a 300 mm quantum wafer foundry, meaning it manufactures the wafers on which quantum devices can be built. 

Crucially, Anderon is intended to provide manufacturing services to customers across the quantum ecosystem rather than only supplying IBM. Quantinuum is approaching the same industrial problem through partnerships. 

GlobalFoundries will fabricate next-generation ion traps and electronics using 300 mm wafer technology, while Monarch Quantum plans to develop and manufacture the lasers and optical components needed for Quantinuum's trapped-ion systems. 

The company has also partnered with Quanta Computer, one of the world's major manufacturers of advanced computing infrastructure, to jointly develop hardware infrastructure, systems engineering and manufacturing capabilities for future quantum systems. There's a wider supply chain behind all of this too.

A joint OECD and European Patent Office study found growing concentration and dependencies around some materials relevant to quantum technologies, including industrial diamonds, aluminium oxide and oxometallic salts. The authors caution that many of these materials also serve much larger industries, so the data can't be read as a map of quantum demand alone. 

Still, the direction is becoming clearer. Quantum computing is starting to develop the less glamorous industrial layer that sits behind almost every mature technology: foundries, component manufacturers, specialist suppliers and companies whose expertise isn't quantum physics at all.

Vendors Are Already Making Different Bets On How To Scale

What's less clear is how much of that industrial capability quantum companies will eventually want to own themselves. IonQ has taken one route. Its acquisition of SkyWater closed in July, bringing semiconductor foundry, advanced packaging and manufacturing capabilities inside the wider company. 

IonQ describes the move as vertical integration, giving it greater control over the path from design through manufacturing. Quantinuum's recent agreements point towards a different model. Instead of bringing each capability in-house, it's working with established specialists across semiconductor fabrication, photonics and large-scale computing manufacturing. 

And Anderon introduces another possibility: specialist quantum manufacturing capacity that can be shared across multiple customers. It's far too early to know which model will work best, or whether different quantum architectures will favour different approaches. But these choices will eventually shape more than production costs. 

Vertical integration can offer tighter control over design and supply. Partnerships can give quantum companies access to manufacturing expertise they don't need to recreate internally. Shared foundries could spread the cost of highly specialised infrastructure across a larger market. Each model also creates different dependencies. 

For enterprises watching the quantum market, the important signal isn't which strategy looks strongest today. It's that quantum companies are beginning to make industrial decisions alongside scientific ones.

Industrialisation Doesn't End At The Factory Door

There's another easy mistake to make when thinking about quantum manufacturing. The job isn't finished when the machine leaves the factory. A commercial system has to reach its destination, be installed, commissioned and kept running. Hardware will need servicing. Components will fail. Systems will need upgrades. 

Customers will need support from people who understand equipment that may operate at extreme temperatures or depend on highly specialised optical and electronic systems. We're already seeing early versions of this downstream model. IBM and Lockheed Martin announced in September that an IBM Quantum System Two will be installed at the Swiss National Supercomputing Centre at ETH Zurich. 

IBM is expected to operate the machine under a three-year arrangement once deployment is complete. The significance here isn't the computing architecture itself. It's the service model around the machine. As locally installed quantum systems become more common, vendors will need to reproduce that capability across more sites.

Installation, maintenance, upgrades and lifecycle support all become part of scaling the technology. A company that can manufacture ten machines but can only support two of them hasn't really solved the industrialisation problem.

What Industrialisation Changes For Enterprise Quantum Strategy

For most enterprises, none of this means they need to start evaluating quantum foundries. It does mean the signals they use to judge the industry's maturity can become broader. Processor performance will remain important. So will error correction, useful applications and the eventual economics of quantum computing. 

But those measures don't tell technology leaders whether the industry developing around the machines is becoming dependable. Manufacturing adds another set of questions:

• Can a vendor reproduce its hardware consistently?
• Is it developing a stable product platform, or does each generation require extensive custom engineering?
• Are critical components available from multiple suppliers?
• Does the vendor own key manufacturing capabilities, or depend on partners?
• If it depends on partners, how concentrated are those dependencies?

Then there are the questions that appear after deployment:

• Can systems be upgraded without replacing everything around them?
• Who maintains the equipment?
• How quickly can failed components be replaced?
• Does the supplier have the people and infrastructure to support a growing installed base?

These aren't questions enterprises need answered before experimenting with quantum through cloud services today. They're signals to watch as the market moves towards more mature commercial deployment. 

QED-C's 2026 industry report helps explain why this view is becoming useful. It counted 556 pure-play quantum companies at the end of 2025, up 8 per cent year on year, while estimating the combined quantum computing and sensing market at $1.9 billion. 

It also identified supply chains as one of the issues increasingly occupying enterprise leaders, investors and policymakers. The industry is growing. The harder question is whether the industrial ecosystem underneath it can grow with it.

Final Thoughts: Quantum Becomes An Industry When It Can Build The Next Machine

The scientific race in quantum computing isn't slowing down. Researchers and vendors still have difficult problems to solve before fault-tolerant systems can deliver the capabilities their roadmaps promise. But another race is now happening alongside it. 

Quantum companies are building foundry capacity, shortening fabrication cycles, working with semiconductor manufacturers, developing specialist supply chains and bringing established computing manufacturers into their ecosystems. Governments are funding some of the industrial infrastructure needed to support that transition. 

None of this proves that large-scale quantum computing has arrived. It shows something more specific: the industry is starting to work on the practical problem of what happens if the science succeeds. Because eventually, building an extraordinary quantum computer won't be enough. 

A mature quantum computing industry needs to manufacture the components, assemble the systems, install them, maintain them and then be able to do it again. And again. For enterprise leaders trying to understand where quantum computing really is on its journey towards commercial maturity, the processor will remain important. 

But the factory, the suppliers and the people who keep the machines running may tell us just as much about what comes next. As those quieter signals become clearer, EM360Tech will continue tracking what they reveal about quantum computing's move from scientific progress towards dependable enterprise technology.