Universal Quantum: Shaping the Future of Fully Scalable Trapped Ion Quantum Computers

While quantum startups keep announcing quantum computers with increasingly many qubits, the reality is that these qubits are error-prone, diminishing what their quantum computers can do today.

To harness the power of quantum computing, it is crucial to increase the number of error-free qubits—a challenging task, especially since there is a trade-off between reducing the number of errors and scaling a quantum computer.

Universal Quantum was founded by Sebastian Weidt and Winfried Hensinger in late 2018 to build fully scalable, universal quantum computers. Using well-established microwave and microchip technology, it uses trapped ion qubits at a more accessible temperature of -200°C compared to other quantum computers that require near absolute zero temperatures and extensive cooling. In addition, its unique modular approach with proprietary interconnects gives it a promising shot to scale quantum computers to millions of qubits.

Universal Quantum raised a $4.5 million seed round in 2020 from Hoxton Ventures , Village Global , Propagator Ventures , Luminous Ventures , and 7percent Ventures , extended by Quantum Exponential in 2022.

This interview is special and longer than usual. It’s based on our first Future of Computing podcast, recorded with Sebastian Weidt, the co-founder and CEO, at this year’s Latitude59 conference in Tallinn, Estonia, and edited for clarity.?You’ll find a recording of the whole interview at the bottom.

Why Did You Start Universal Quantum?

I don't have a story about being five years old and starting to dream about quantum computing. But I've always been passionate about working on things that have an impact and that can move the needle for society.?

Originally, I planned to go into business, but I decided first to pursue a degree in physics and return to business later. After my undergraduate studies, I made a brief foray into management consulting and then did a PhD in quantum physics. It’s when I realized that quantum computing is not merely an academic endeavor but will have commercial applications in the future. And, it will be difficult for anything to beat quantum computing’s impact in the long term, which excited me a lot.?

But it had to be done at scale and, therefore, as a business. The founding of Universal Quantum closed the circle for me as I got back into business, specifically to bring quantum computers to the market.?

What Exactly Is Quantum Computing?

Quantum computing is a new technology for solving some very difficult computational problems that even the most powerful supercomputers and AI can’t solve today.

Unlike many attempts to advance computing, quantum computing is not about keeping Moore's law alive for a while longer. It's about solving problems that would take any classical supercomputer unreasonably love, like a million years. Quantum computers excel at tackling those kinds of problems.?

Their fundamental building blocks are quantum bits, in short qubits, analogously to bits in a classical computer. However, while bits are either one or zero, in the quantum world, qubits can be one and zero simultaneously. In the quantum world, weird things are constantly happening—an atom can be in two places simultaneously, and we get effects like entanglement that are pretty mind-boggling. Quantum computers use these quantum effects to compute things in a fundamentally new way.?

Why Are You Using Trapped Ions for Building a Quantum Computer?

To build a quantum computer, you need a physical system that can experience quantum phenomena to serve as a qubit. This could be a photon, an atom, a trapped ion, or a superconducting qubit, which is why quantum computers can look quite different.

The actual question is: What does a quantum computer need to look like to give us utility and unlock commercial applications? You need a quantum computer with many high-quality qubits that you can control well at scale. And trapped ions tick many of these boxes: Currently, they make for the best qubits and the best-performing quantum computers.?

Reaching scale is one of the industry's key problems and also a common concern for ion trap quantum computers: How do we go from the small-scale quantum computers we have today, which are great toys but commercially useless, to large-scale quantum computers that can solve commercial problems?

The best quantum computers today have tens or maybe hundreds of qubits. But, we will need to get to millions of qubits. Many people in the industry don't have good answers. And that’s why we focused on this question for many years. Trapped ions provide a quantum hardware platform with many nice properties, and we believe we found a way to scale it.?

How Do You Evaluate Progress in Building a Quantum Computer?

Quantum engineers often talk about gate fidelity, the quality of quantum operations, which indicates how well you can process information. But they often don't talk about how well a quantum platform can scale. That’s hard to quantify, and I don’t have a simple metric either; you need to dive into the technology and understand the roadblocks to scaling up.?

However, there's broad agreement in the industry that large-scale quantum computers will require a modular approach. That means each module needs to act as a mini quantum computer, and you must be able to interconnect them. In classical computing, we already use multiple cores, but it’s incredibly hard to connect quantum chips in a quantum way. There’s no industry standard solution. Yet, we have found a way forward.?

How Do You Connect Quantum Modules?

When we talk about interconnects within quantum computing, there are two regimes: One is where you're connecting quantum chips or quantum modules within a quantum computer. That’s where people often point to photonic interconnects, but they work rather poorly. The other one is connecting different quantum computers across the globe, e.g., between London and Tallinn—basically, when building the quantum internet. And that’s where photonic interconnects play a crucial role.?

When you're within a quantum computer, operations have to happen rapidly, and the quality of components matters a lot. These are major challenges for photonic interconnects, and that's why we've worked hard to eliminate the need for them.?

About a year ago, we published a paper demonstrating a link between two quantum modules that is as pure as it gets. It’s fast and has a high fidelity—a near-perfect link. Many people are looking into connecting quantum modules but often need to add another platform, like photonics, to get these interconnects right. Having that out of the box with very high fidelity, thanks to our own interconnect technology, UQ?Connect, is a big deal and opens up the opportunity to scale up our quantum computers.?

How Important Is Quantum Error Correction?

Quantum states are very fragile, and one of the major challenges in quantum computing is protecting them from their environment. Otherwise, the environment may introduce noise and disturb their quantum state, which leads the quantum computer to become classical, i.e., it destroys its quantum nature.?

As it turns out, we will never have a perfect qubit, as you can’t avoid noise entirely. This means we must find a way to programmatically remove the errors in a quantum computation that stem from noise.?

Similar to conventional computing, where error correction algorithms remove bit-flip errors on the transistor level, we need quantum error correction protocols. What they say on a high level is: give me lots of noisy, error-prone qubits, and I remove these errors for you and give you a smaller number of very pristine, error-free qubits, also called logical qubits. These are the ones that you can actually use to process quantum information.?

That's where that big goalpost comes from: We have to push the number of qubits up dramatically, as we’ll need many for error correction. It’s a tough challenge for every quantum startup. However, some may see this as an unsurmountable challenge for their platform, so they’ll focus more on what we can do with the noisy, error-prone quantum computers that we have in the near term—called the noisy, intermediate-scale quantum (NISQ) era.?

Quantum engineers worldwide are doing remarkable work trying to figure out what current quantum computers could be capable of. However, I don't think we’ll see utility from quantum computers without error correction. That’s why we have always focused on scaling up the number of qubits. Even if we find certain applications for smaller quantum computers, we will always want to have more qubits and, thus, more computational power to unlock the next application. Focusing on scaling up is the way forward.?

Up to What Point Do You Optimize Individual Qubits, and When Is It Time to Scale Up?

The total number of qubits we’ll need for quantum error correction depends greatly on their fidelity—the better quality qubits you have, the lower the number of physical qubits you need. That’s why improving the fidelity is so important.

However, from what we're seeing in the field, you can’t quite switch from a few nearly perfect qubits to scaling the number of qubits out of the box. When people built some of the early quantum computers, they focused on getting five- or ten-qubits to work perfectly. However, once they wanted 100 qubits, they had to re-engineer the entire quantum computer, which also impacted the fidelity.?

That’s why we don't think it's wise to focus on getting just a couple of perfect qubits. We approach it the other way around, thinking carefully about what a million-qubit quantum computer must look like and then working backward to see what we need to do today.?

That way, we end up front-loading some of the difficult engineering tasks. But as we tackle them early on, we’ll get a quantum computing platform that can scale naturally. All the core engineering blocks needed for a million-qubit machine are already in our smaller-scale quantum computers, such as the ones we're building right now, for example, for the German government. It will allow us to scale up much faster than otherwise possible.?

How Many Qubits Will We Need To Reach Quantum Utility?

While the target is to push towards millions of physical qubits, we likely won’t need millions of logical qubits to do something useful. For example, breaking RSA encryption, which would impact national security, is one of the toughest challenges, but it would require just a couple of thousands of logical qubits, depending on their quality.

NISQ computers have been useful?because people can go online today and play around with them. They don’t deliver commercial value for solving problems just yet, but engaging with the technology is important in many ways. It gets engineers excited to become quantum engineers and developers to become quantum software developers.?

Also, companies need to understand what quantum computers can do and what this will mean for their business models. Quantum computing is so fundamentally different from traditional computing that it takes time to prepare and acquire the necessary knowledge to harness it once quantum computers become ready and widely available.

Do You Optimize Quantum Computers for Certain Applications?

We focus on developing universal quantum computers to run any quantum algorithm. That said, how efficiently you can run a particular quantum algorithm depends on your quantum computer’s connectivity. Implementing steps in a quantum computation, called quantum gates, is similar to logic gates in classical computers and requires qubits to communicate. A quantum algorithm might require qubit number five and qubit number 2050 to communicate, and the question is, how do you implement that?

We have a fully connected quantum computer that can address those demands with little overhead. Other quantum hardware platforms arrange qubits in fixed places, so you can't move them physically, and if you want to make distant qubits talk to each other, you must perform some very error-prone operations. We can physically move our qubits in space and get them close to other qubits so they can exchange information—and moving our qubits is a remarkably error-free operation.?

That's where our fully connected architecture will play to its strengths—we can change the connectivity of our system through software. If an algorithm demands different connectivity, we can just program it and don’t need to touch the hardware.?

What Applications Are You Most Excited About?

One of the main reasons I started Universal Quantum is to do something good and positive for society. And few things could have such a big impact in the long term as quantum computing as it will affect nearly every sector, be it security, material science, or finance.?

I’m particularly passionate about applications in healthcare, where it will have an impact across the board for logistics, medical imaging, and drug development. Take drug development as an example: currently, it takes more than ten years to develop a new drug. That’s among others, because it's tough to understand molecules and chemical reactions. They are quantum by nature, and classical computing is inefficient at modeling such quantum problems. However, some quantum algorithms could solve them more straightforwardly, becoming a handy tool for pharmaceutical companies to develop drugs faster and benefit the healthcare sector.

Will There Be a ChatGPT Moment for Quantum Computing?

A ChatGPT moment for quantum computing will come when we see the first utility of quantum computers. At Universal Quantum, we see it as our mission to provide society with quantum computing as a tool to do awesome things. We won’t develop new drugs ourselves, but we’ll provide the computational resources for scientists to do so. And we’ll see quantum computing enable all kinds of applications.?

People won’t use Quantum computing on their phones daily like ChatGPT. But they will experience it indirectly, as it will enable things, like developing a new drug or material, that wouldn’t have been possible without it.

We're still beginning to uncover what quantum computing can do for society. We already know enough to get excited about it and for governments and investors to invest in it. Think about how far conventional computing has come—no one could have imagined the incredible advancements we see today. Quantum computing is on a similar trajectory.?

How Do You See Progress in Quantum Computing Globally??

There are huge government initiatives for quantum computing worldwide right now; some even say there is a quantum race. It leads to a great push on a national level and a lot of funding being made available, which is important to push the sector forward.?

Still, people involved with the sector have to be aware that it’s a marathon, not just a single sprint. We can’t expect everything to happen in the next couple of years, but we need a long-term view, and then quantum computing will be quite transformational.

It’s no secret that European startups don’t often receive the same level of funding as those in the United States. Especially for NISQ, the United States plays a major role because capital was available to push those players forward.?

However, what is promising to see, especially in Europe, is not just the strong academic foundation for quantum physics but also the increasing number of spinouts in quantum computing and quantum technologies more broadly. There’s a great opportunity to harness our diversity and talent in Europe to be at the forefront of one of the next major technology waves.

Where Will Universal Quantum Be in Three Years?

You will look at a company that has successfully introduced quantum computers to the market. You could approach us with inquiries such as, "You have X number of qubits here; can you provide quantum computers with ten or a hundred times that amount?" Our response will be, "Yes, we can do that."?

Looking at how the number and quality of qubits have increased over the years, it seems linear and almost incremental. If we keep going like this, reaching an order of magnitude that matters will be tough, so we need a more exponential scaling of high-quality qubits. We’re working on a step change in scaling the number of qubits with our platform, and we have a good idea of how this will lead to business value.?

How Will Customers Access Your Quantum Computers?

Currently, many quantum startups are making their computers available via the cloud so that you can access them easily on your computer. But there are good reasons?why people purchase quantum computers and install them on-premise.

A few years ago, we sold two quantum computers to the German government for about €70M. We’ll deliver these machines to their facilities, and they’ll have complete ownership. That makes a lot of sense in certain scenarios where people want local sovereign capabilities and don’t want to rely on cloud access to quantum computers located who knows where. We give access through the cloud and on-premise installations.?

What Advice Would You Give Fellow Deep Tech Founders?

Especially in deep tech, you have to keep going. It's often not as sexy as many investors would like in the short term. It's hard from a technical perspective; you're constantly pushing things. And that excites me, but it is still hard. You've got to have the conviction and the long-term drive to make your vision a reality.?

The other thing about deep tech startups is that the founders are very technical. And as technical people, we love our tech and get excited about things that the outside world doesn't quite appreciate or get excited about. So, a lot of work has to go into finding a way to communicate your technology so that other people can understand it and get excited.?

We’re actively working on that. I would always advise other technical founders not to underestimate this because what you might find remarkable from a technical perspective might not interest others. Think about how it changes their lives. Communicating that properly requires a lot of work.?

How Do You Build a Strong Team?

One thing that concerned me initially was how well the hiring side would go. As a deep tech startup, we rely on heavy engineering and need more engineers than physicists. The founding team consisted of physicists, and we've done a lot of research on scaling quantum computers. But to make it happen, you need smart engineers.

How do you pull the world's best engineers from the Googles and Apples to come and work at a little startup when you can't compete on salary and benefits? It worried me in the beginning, but it turned out not to be a problem at all. It’s quite the opposite: all the smart and passionate people want to work at the cutting edge—it's not all about money and benefits; they want to work on something that can change the world.?

Who Should Reach Out to You?

Given our current contracts, we're focused on developing our tech and are strongly revenue-driven. That said, we are always open to speaking with people who are passionate about what we're building, particularly people who align with our vision for quantum computing, have a longer-term mindset, and an appreciation of where we need to go.