Google captivated the tech world Monday with Willow, a new quantum chip that outperformed even the world’s best supercomputer on an advanced test, and experts say this week’s revelations are just the start of a new phase in cutting-edge computing.
The new chip, which wowed even Tesla CEO Elon Musk, can complete in five minutes a complex computation that would take the most powerful supercomputer 10 septillion years—more than the estimated age of the universe, wrote Hartmut Neven, the founder and leader of Google Quantum AI, in a blog post.
Google researchers were also able to prove for the first time that the chip’s errors did not increase proportionately as the number of qubits, the basic information processing unit in a quantum computer, rises. The qubits in quantum computing are more prone to mistakes than a regular computer because they are highly sensitive to electromagnetic and other forms of interference and can only remain in a quantum state for very brief periods of time.
But in order for a quantum chip’s calculations to make sense, it needs to have a very low error rate, Javad Shabani, a physics professor and director of the Center of Quantum Information Physics at New York University, told Fortune.
Google’s breakthrough is “one of the highlights of the recent decade,” he said, and could mean quantum computers are a step closer to being used for more practical purposes.
“Classical” supercomputers, including the Frontier supercomputer, which is commonly recognized as among the most powerful, are used for advanced modeling and simulations. Yet, these supercomputers are more similar to the common laptop than you’d think. They operate on the binary system of 0s and 1s that less powerful computers use, albeit at a much more advanced level. While supercomputers are undoubtedly powerful, they are huge pieces of machinery that use massive amounts of energy for cooling systems.
Quantum chips like Willow operate on a much smaller scale. The qubits that make up these high-powered chips are roughly the size of an atom, and operate via the rules of quantum mechanics. The chips pack a computing punch at a minuscule level and with potentially better energy efficiency than supercomputers, but they’re also limited in their uses.
The test Google ran on Willow is a “benchmark” meant to show the quantum chip’s capability relative to other systems, including the best existing supercomputers. Google in 2019 completed a benchmark test with its earlier Sycamore quantum chip and found at the time that the older chip could complete a calculation in 200 seconds that would take the best supercomputers thousands of years to complete, according to a blog post at the time.
Google’s Willow chip has around double the qubits used to power its computing than that of the Sycamore chip, and exponentially more computing power than a supercomputer has for some uses.
Google’s Willow findings show that quantum computers can perform much better than the best available supercomputers in certain cases, an assertion that just a few years ago was more widely doubted. It also possibly shifts the focus of future quantum chip development away from primarily trying to find a way to reduce errors to, instead, trying to increase the number of qubits that make up each chip, according to Rice University physics and astronomy professor and founder of the Rice Quantum Initiative Kaden Hazzard.
Google’s discovery is groundbreaking, but that doesn’t mean your standard laptop will be replaced by a quantum computer anytime soon, Hazzard told Fortune.
“It’s at least a few years away before we reach the kind of capabilities where the quantum computer can be used directly for things that have business applications, for example,” Hazzard said.
Fears about Willow having the capacity to decrypt even the most advanced encryptions used in banking or tech are also overblown, Rice University assistant professor of computer science Tirthak Patel told Fortune. Solving such advanced cryptography would require a much more advanced chip than Willow, said Patel, and researchers are already finding ways to create quantum-proof encryptions.
Instead, Willow is a stepping stone toward greater progress in quantum computing. Monday’s findings confirmed it outperforms supercomputers in some cases and will do lots to convince naysayers of the potential quantum computing has to eventually help improve human lives, Patel said.
Still, while it may be several years away, advanced quantum computing could be used to accurately simulate minuscule quantum systems like chemical reactions that could help discover new drugs, or solve quantum mechanical problems to build new types of materials or better catalysts for reactions, said Hazzard.
Possibly sooner, the advanced computing ability of quantum chips could help improve weather predictions or streamline supply-chain efficiency, all while being potentially less energy-intensive than existing supercomputers, Patel said.
Because these types of applications can already run on quantum chips with fewer qubits and without the advanced error reduction that Google has shown is possible, better quantum chips could revolutionize these areas, said Patel.
“The biggest task right now that most scientists and engineers working in the quantum computing field are concerned with is demonstrating the utility and practicality of quantum computing for different real-world applications,” Patel said. “While this experiment hasn’t done that just yet, it is a major milestone in that step.”
