Huge fault-tolerant quantum computers on the agenda at Commercialising Quantum 2024 conference

15 Jun 2024 Hamish Johnston

Collected at: https://physicsworld.com/a/huge-fault-tolerant-quantum-computers-on-the-agenda-at-commercialising-quantum-2024-conference/

On 5–6 June I was in the City of London for Economist Impact’s Commercialising Quantum 2024 conference. Held in a shiny new skyscraper in Bishopsgate, the technology being discussed felt as new and exciting as the venue itself. As someone who was taught quantum mechanics well before the second quantum revolution, I felt more akin to the ancient churches that I walked past to get there – a vestige of a different era.

I was at the conference last year, so I was keen to see how the commercialization of quantum was proceeding. Was it rapidly rising to the heavens like a modern skyscraper, or was construction proceeding at a more leisurely pace like a medieval cathedral?

One company that has taken great strides since June 2023 is US-based PsiQuantum, which earlier this year received a staggering A$940m ($620m) from Australian federal and state governments to build “the world’s first utility-scale quantum computer” in Brisbane.

The company’s founder and CEO Jeremy O’Brien spoke at the event and I was very keen to find out how the company was scaling-up the production and integration of its photonic quantum chips. The Brisbane facility will need about one million physical qubits so that quantum error correction can be used to create the thousand or so logical qubits required to create a fault-tolerant computer. The ultimate goal being to use this system to do certain calculations that are beyond the means of even the most powerful supercomputers.

Mass production

O’Brien, who is Australian, said that PsiQuantum is using technology first developed by the semiconductor industry to mass produce their chips right now. He said that integration will be key to the company’s success and using photonics to connecting qubits using light is the way forward. He described the system as a hybrid of photonic and electrical components and said that it will have to be cooled to cryogenic temperatures because of its superconducting photon detectors.

He suggested that the system will be up and running in a little over three years – which seems very soon to me! Incidentally, it took about three years to build the skyscraper (22 Bishopsgate) in which O’Brien spoke.

Echoing O’Brien’s assertion that millions of physical qubits are needed to create the 1000s of qubits required to do useful calculations was Oded Melamed, who is CEO of Quantum Source. The Israel-based company is designing a quantum computing system that integrates atomic qubits – which Melamed said are entangled easily – with photonic qubits, which are much more easy to connect up. The atomic qubits comprise rubidium atoms, which are stored in small vacuum cells and coaxed out individually to interact with photons in optical resonators.

Quantum Source is a much newer company than PsiQuantum, which was founded in 2015. The Israeli company came into being in 2021 with $27 million in seed funding.

Quantum factories

Melamed referred to large, integrated implementations of millions of qubits as “quantum factories”. They would be huge – possibly the size of a football pitch. When you connect millions of logical elements – quantum or otherwise – together, it will be possible to do calculations. However, a debate I can see looming on the horizon is whether calculations done at such facilities are actually quantum rather than classical.

Error correction is required because the quantum nature of most types of qubit is very quickly degraded and destroyed in a process called decoherence. A big problem with current error-correction schemes is that they use a very large number of physical (real) qubits to create one logical (virtual and useful) qubit. Today, that ratio is greater than 1000 to one. When you consider that useful quantum calculations (those that far outperform conventional computers) will require thousands of logical qubits, a quantum computer that uses error correction would need millions of physical qubits – as O’Brien and Melamed pointed out.

But what if the number of physical qubits could be reduced significantly? That was at hinted by the physicist Peter Knight, who advises the UK government on all matters quantum. He told the conference that great strides have been taken in the development of error correction schemes that are less demanding in terms of physical qubits. However, he also pointed out that it is possible that the effects of some types of noise – the absorption of cosmic rays, for example – may always plague quantum computers. This is one reason why the UK government is looking at expanding the Boulby Underground Laboratory – to provide low cosmic-ray background facility for developing quantum devices. And, maybe even running a quantum computer more than 1 km under the North Sea.

Keeping an eye on quantum

Many of the delegates that I spoke to at the conference did not work for quantum-technology companies – but rather work for large companies such as Nestlé and Johnson & Johnson. These people are charged with understanding what quantum computers could do for their companies in the future.

Indeed, many of these companies have gone beyond the observing phase and are actively engaging with quantum technologies. I attended a fascinating talk by the physicist Lene Oddershede of Denmark’s Novo Nordisk Foundation – which (amongst other things) is the majority voting shareholders in the pharmaceutical giant Novo Nordisk. She said that the foundation is three years into its “quantum mission” to support the development of quantum simulations of natural systems and quantum sensors for biomedical applications. To this end, the for-profit foundation has created a company called Quantum Foundry Copenhagen with the sole function of protecting the intellectual property developed during the quantum mission. She said that because of Danish law, universities would not be able to protect this IP.

What bad actors could do with quantum technologies is a concern of Matija Matokovic, who is deputy head of innovation at NATO. He described quantum technologies as disruptive and emerging technologies with the potential to change the nature of security. He also pointed out the strategic importance for countries to develop and nurture quantum technology companies that are developing strategically important technologies.

Few physicists

Oddershede, Knight and O’Brien are all physicists – and that put them a distinct minority at the conference. I spent some time at the Institute of Physics’ exhibition stand chatting with delegates (the IOP publishes Physics World). The first question I normally asked was “are you a physicist?” and the answer was almost always “no”. That came as no surprise because the focus of the conference was about commercialization.

Many of the talks that I attended were not given by physicists (or other technical experts), but rather by start-up CEOs, who are often serial tech entrepreneurs. While I cannot claim to be an expert in quantum computing, I did find it much easier to follow talks at the conference that were given by technical experts. And, the presentations that I struggled with the most tended to be discussions about the future benefits of quantum computing.

One puzzling example – and I will give a general description because I don’t want to cast aspersions on the speakers, whom I believe were sincere – was a fireside chat between the representative of quantum computing company and one of its customers. The customer was talking about the success that they had in showing that a quantum algorithm could be used to solve a business-related problem. But, it wasn’t clear whether this quantum algorithm was run on a quantum processor or on a conventional computer that was simulating a quantum computer. To me, this is an important distinction that may have been lost on non-technical audience members.

While most of the focus of the conference was on quantum computing, there was much said about quantum sensors – which are often much more mature technologies than quantum processors. I watched a fascinating presentation by Margot Taylor, who is director of functional neuroimaging at Toronto’s Hospital for Sick Children and David Woolger, who is CEO of UK-based Cerca Magnetics. Cerca makes a brain scanner that uses quantum sensors and Taylor described how she is using it to study cognitive development in children.

Avoiding hype

While the science, engineering and business strategies of the future discussed at the conference were fascinating, it was easy to get swept away by the hype – something that speakers including Peter Knight warned against.

However, there were also some real success stories. My favourite comes from Cerca’s David Woolger, who pointed out that his company was formed in 2020, and has been in profit ever since.