A new wave of confidence — and capital — is sweeping through the quantum computing industry.
The governments of Australia and the state of Queensland this week committed A$940mn (US$620mn) between them to back the construction of a full-scale quantum computer near Brisbane by US start-up PsiQuantum. The deal was just the latest sign that a decades-old dream of a new form of computing that takes advantage of the unusual properties of quantum mechanics may finally be coming to fruition.
The system in Australia will be “the first machine that crosses over the threshold into [being a] really useful computer — the first quantum system in the world that will be commercially useful”, claimed Pete Shadbolt, PsiQuantum’s chief scientific officer.
He is not alone in making grand claims for a generation of computers that are scheduled to be built before the end of this decade, signalling a relatively short sprint for a field that has had its share of setbacks in the 65 years since professor Richard Feynman first laid out the idea of quantum computing.
One of the biggest challenges has been that the quantum bits — or qubits — used in today’s machines are highly unstable and only hold their quantum states for extremely short periods, creating “noise”. As a result, faults accumulate during any quantum calculation, making the computer essentially useless.
Recent advances in error correction, a technique for encoding information into qubits that compensates for this, have promised a way past this problem far sooner than most in the industry had expected.
Other companies that have set their sights on a new finish line in the quantum race include IBM, which has been building experimental quantum systems for years. In late 2023 it laid out a road map for the first time to reaching a fully functional, practical system.
“I feel like we have a path to scaling to demonstrate a fault-tolerant quantum computer” by 2029, said Jay Gambetta, vice-president of quantum computing at IBM.
“No one’s ever integrated a system with millions of qubits before, the most is a few thousand,” said Scott Aaronson, director of the Quantum Information Center at the University of Texas.
For years, predicting exactly when quantum computing will reach practical utility has seemed like a fool’s errand. Hopes of being able to program an earlier generation of so-called noisy machines — systems with unstable qubits — have been dashed, ending one avenue to practical use.
And even some seemingly significant research breakthroughs have failed to yield results. When Google claimed in 2019 that it had reached “quantum supremacy” — the point at which quantum systems leap ahead of traditional computers — it turned out that new ways of programming existing machines could erase the advantage.
But many in the industry now believe that in recent months, a clear route has opened up to building large-scale systems that will bring real technical and business advantages.
A series of research breakthroughs, starting with Google last year, has brought major advances in error correction years earlier than most expected. Research results from Harvard University and Boston-based group QuEra late in 2023, along with a paper this year from Microsoft and US- and UK-based Quantinuum, have added to the hopes.
“Over the past year, we’ve seen a large amount of progress in the industry, and in particular around error correction,” said Steve Brierley, head of Riverlane, a UK quantum start-up. That has reduced the number of technical breakthroughs that still need to be made, he said. “It increasingly seems less like a science problem and much more like an engineering problem.”
One result has been a new race to scale up today’s quantum hardware systems to be able to handle the useful algorithms that companies such as IBM now believe will be soon be produced.
Australia’s investment in PsiQuantum, in the form of equity in the company as well as grants and loans, marks an unusual bet on qubits based on photons, rather than the superconducting qubits used by IBM and Google. Photons have been shown to provide a stable base for a quantum system, but PsiQuantum has in the past published little research to show what progress it has made to overcome the many hardware challenges, said UT’s Aaronson.
In an effort to lay those doubts to rest, PsiQuantum this week published its first research paper outlining hardware advances that Shadbolt claimed had put it well on the path to a workable, large-scale system by the end of the decade.
With attention turning to the practicality of manufacturing large numbers of qubits and quantum chips, as well as other hardware needed to link them together in larger systems, PsiQuantum also claims advantages for photonic technology that will get it to a 1mn-qubit system faster than others in the industry.
In one positive sign for the industry as it moves towards commercial manufacturing, a more developed supply chain in quantum hardware has been forming as the technology moves out of the lab, said Brierley.
Despite the new hopes, experts such as IBM’s Gambetta admit that their plans to get to full quantum computing could still be derailed. Predicting when necessary algorithmic breakthroughs will be made to make error correction practical is harder to predict than creating a road map for scaling up hardware, he said.
Gambetta added, though, that if IBM could show the first, very early practical results from quantum computing in the next two years — a point it calls quantum advantage — it could trigger a new wave of interest from the corporate world.
“People have these plans on paper — the things the experimentalists predict always take longer than they think,” said Aaronson. But after recent advances, the nascent quantum computing industry seems for the first time to have a real shot at reaching practical usefulness, he added.
