In the race between information companies like Google, IBM and Intel to create the fastest quantum computer, a report found by the Financial Times says that Google has finally pushed ahead of the rest and reached “quantum supremacy” — for now at least.
To achieve “quantum supremacy”, a quantum computer must be capable of solving a complex problem that would otherwise be impossible for a classical computer to solve in its lifetime.
According to the report, Google’s ‘Sycamore’ was able to do this by completing a complex calculation in just three minutes and 20 seconds, compared to the estimated 10,000 years it would take the world’s most advanced classical computer, Summit.
The paper was published on the NASA website before it was taken down, FT reports. The paper has resurfaced online, though.
Just as a classical, or non-quantum, computer makes calculations using pieces of information called bits that can either represent a zero or one, quantum computers also use binary units, called qubits, that can simultaneously represent both zero and one. This duality gives quantum computers the potential to be much, much faster than their classical counterparts. More importantly, the ability of quantum computers to crunch huge amounts of data in a short amount of time would be incredibly useful for advancing data heavy fields like pharmaceuticals, machine learning and even finance and agriculture.
“This dramatic speedup relative to all known classical algorithms provides an experimental realization of quantum supremacy on a computational task and heralds the advent of a much-anticipated computing paradigm,” the paper reportedly says.
To complete these dizzying calculations, New Scientist reports that Sycamore’s system contains 54 quantum entangled superconducting qubits. While only 53 of the qubits reportedly worked during the test, this still outpaces IBM’s upcoming 53-qubit and Intel’s 49-qubit machines.
While some are heralding this as the beginning of a new quantum age, other researchers have less eager to ascribe such laurels to the results just yet, particularly IBM’s head of research, Dario Gil, who said to the Financial Times that it’s too early to celebrate this milestone, especially because it’s likely that Sycamore was operating within a very narrow question space. In other words, it’s likely that Sycamore was specifically trained to complete the given problem and to complete it well, meaning that despite its computing power it’s not necessarily a very flexible machine.
But, if Sycamore’s accomplishment is valid it would be a very exciting moment for the quantum computing research community and could be a first step in creating scalable quantum computers that could be used more widely, an engineering professor at the University of Southern California, Daniel Lidar, told the Financial Times.
By successfully reducing the amount of “cross-talk,” or interference between qubits, Sycamore was able to control for errors other machines can not, said Lidar.
“They have demonstrated a path to scalable quantum computing,” Lidar told the Financial Times. “Once you have a fully error-corrected quantum computer, the sky’s the limit.”
While these accomplishment won’t translate to personal quantum computing anytime soon, that doesn’t mean there wouldn’t be tangible benefits if quantum computing were to become widely used.
Instead of improving the speed your Twitter feed refreshes, a 2018 Boston Consulting Group report says that quantum computers would be implemented in research labs that deal with massive amounts of data, including scientific fields like cryptography, material science, pharmaceuticals and machine learning, as well as logistics, manufacturing, finance, energy and agriculture.
Yuri van Geest, founder of SingularityU in the Netherlands and an expert in the singularity, said in 2016 that more creative, liberal-arts driven jobs that analytical machines struggle to accomplish may be on the rise as human technicians become less essential in other fields.
Classical computers aren’t going anywhere anytime soon. But, these new improvements in quantum computing represent yet another brick in the road towards being able to do important science more quickly with the help of these number-crunching beasts.