Google’s quantum computer solved a problem that took 10 septillion years in five minutes

The Willow processor contains 105 qubits, which serve as the fundamental building blocks of quantum computers. Qubits have high performance, but are error-prone as their performance is affected by tiny noises.

In a paper published Monday in the journal Nature, Google said it has found a way to combine qubits in a way that reduces error rates while increasing the number of qubits. The company also said it has learned how to correct errors in real time, a key step toward creating practical quantum machines.

So far, the mathematical problems that quantum computers solve have no practical application, and the propensity of computers to make errors makes them of little use in areas that require precise calculations. However, the progress Google engineers have made is approaching the point where quantum computing can be used outside laboratories, Google Quantum AI founder Hartmut Neven said in a conversation with Bloomberg.

According to him, the company intends to present a real example of the use of a quantum computer next year. “This is now possible,” the scientist stressed.

The race to master quantum computing is taking different approaches. Google’s technology is based on superconducting qubits, which are also used by the company’s competitors such as IBM and Amazon. At the same time, Google previously invested in the company QuEra Computing, which develops technologies based on so-called neutral atomic qubits.

“When we decide to scale, we want to be absolutely sure that we choose the most promising technology. We are betting on superconducting qubits. But perhaps QuEra will show us the benefits of neutral atoms. We’ll see,” said the founder of Google Quantum AI.