<img src="https://d1o50x50snmhul.cloudfront.net/wp-content/uploads/2017/06/22165204/image03.jpg">

Google is leading the pack when it comes to quantum computing. The company is testing a 20-qubit processor – its most powerful quantum chip yet – and is on target to have a working 49-qubit chip by the end of this year.

Qubits, or quantum bits, can be a mixture of 0 and 1 at the same time, making them potentially more powerful than classical bits.

And if everything goes to plan, the 49-qubit chip will make Google the first to build a quantum computer capable of solving certain problems that are beyond the abilities of ordinary computers. Google set itself this ambitious goal, known as quantum supremacy, in a paper published last July.

Alan Ho, an engineer in Google’s quantum AI lab, revealed the company’s progress at a quantum computing conference in Munich, Germany. His team is currently working with a 20-qubit system that has a “two-qubit fidelity” of 99.5 per cent – a measure of how error-prone the processor is, with a higher rating equating to fewer errors.

For quantum supremacy, Google will need to build a 49-qubit system with a two-qubit fidelity of at least 99.7 per cent. Ho is confident his team will deliver this system by the end of this year. Until now, the company’s best public effort was a 9-qubit computer built in 2015.

“Things really have moved much quicker than I would have expected,” says Simon Devitt at the RIKEN Center for Emergent Matter Science in Japan. Now that Google and other companies involved in quantum computing have mastered much of the fundamental science behind creating high-quality superconducting qubits, the big challenge facing these firms is scaling these systems and reducing their error rates.

It is important not to get carried away with numbers of qubits, says Michele Reilly, CEO at Turing Inc, a quantum start-up. It’s impossible to really harness the power of these machines in a useful way without error correction, she says – a technique that mitigates the fickle nature of quantum mechanics.

Ho says it will be 2027 before we have error-corrected quantum computers, so useful devices are still some way off. But if Google can be the first to demonstrate quantum supremacy, showing that qubits really can beat regular computers, it will be a major scientific breakthrough.