 © Google Quantum AI
current progress 90%
Richard Waters in San Francisco
Published2 hours ago
9Print this page
Google claimed to have gained an edge in the race to demonstrate the first practical use of a quantum computer, revealing an algorithm that it said could help to open up new techniques in drug discovery and the development of new materials.
The internet giant called its demonstration an example of “quantum advantage” — the point at which quantum systems gain a clear lead over traditional computers.
The quantum computing world has been poised for evidence that quantum machines are finally reaching a point where they can outperform on certain types of calculations.
In a paper published in the science journal Nature on Wednesday, Google described an algorithm that simulates the quantum mechanical behaviour of systems in nature, such as the way the atoms in a molecule interact.
The company said that, by running the algorithm on its latest quantum chip, it had obtained results 13,000 times faster than would have been possible on a “classical” supercomputer.
The results were also verifiable, Google said, meaning they could be replicated on a second machine — something it had not been able to show before, and an important step towards building confidence in the calculations performed by quantum computers.
Hartmut Neven, head of Google’s Quantum AI research lab, called the work a “demonstration of the first algorithm with verifiable quantum advantage” and “a milestone on the software track”, as more research effort is directed towards developing practical algorithms that can run on quantum systems.
 © Tomohiro Ohsumi/Getty Images
However, the company stopped short of claiming the work would, on its own, have practical uses. Instead, it said the algorithm, known as Quantum Echoes, demonstrated a technique that could be applied to other algorithms used in areas such as materials science and drug discovery.
Its work on Quantum Echoes is “applicable in a lot of quantum simulation systems, and so this is a path to practical quantum advantage, something that would be not only better on a quantum computer, but useful,” said Charina Chou, chief operating officer of Google’s Quantum AI Labs.
Recommended
Visual story
Quantum computing could break the internet. This is how
In a sign of where it might yield practical results, a second paper published by Google on Wednesday showed how its method could be applied in nuclear magnetic resonance, a technique that is already commonly used to study molecules. That work involved an experiment on a relatively small quantum system that was not able to work faster than a traditional computer, meaning it fell short of full practical quantum advantage.
Previous claims have been met with scepticism. US start-up D-Wave said it had reached the quantum advantage threshold earlier this year, while a group of researchers from the University of Texas and Quantinuum this month said they had shown a form of quantum advantage. Google itself made a controversial claim in 2019 of a similar breakthrough known as quantum supremacy.
However, other researchers have shown that results like these could have been replicated on traditional computers in far less time, or have questioned whether the types of problems solved in these demonstrations will ever have any practical use.
 © Google Quantum AI
Google’s latest claim has yet to be scrutinised by rival researchers, though it went through a peer review process before publication in Nature. In an unusual move, the company also said it had exhaustively “red-teamed” the research, or putting some of its researchers to work in trying, and failing, to disprove its own results.
Researchers from IBM predicted recently that there were likely to be a number of claims of quantum advantage before the end of 2026, amounting to “a pivotal threshold for the field” as quantum computers come closer to practical use. But it cautioned that the claims would become the subject of intense debate among other researchers and that it was likely to take some time before there was general agreement.
Google‘s controversial 2019 claim to quantum supremacy, also made in a paper in Nature, involved solving a problem in 200 seconds that it said would have taken 10,000 years on a classical supercomputer.
Other researchers quickly showed ways to do the calculation much faster on a classical machine, and Neven said that improvements in chips since then meant it would only take six seconds on the latest GPUs.
But he added that Google’s quantum processors had shown even faster improvement. “This is a race classical machines can’t win,” he said. |
