Future uses of quantum
Building quantum computers isn’t the only challenge—knowing what to do with them is just as important.
A quantum computer will only be useful if it can solve practical problems – and in particular solve them more efficiently than a classical computer. In fact, a quantum computer has already beaten classical computers. In 2019, Google engineers demonstrated that a quantum processor can solve a certain mathematical problem faster than a state-of-the-art supercomputer. While an important milestone, the mathematical problem solved was especially designed to be hard for a classical computer, and of very limited practical use.
Finding and realizing a useful application of a quantum computer is being heavily researched: programming a quantum computer is very different from programming its classical counterpart. We know that a quantum computer will not solve every problem faster than a classical computer – even as quantum computers become larger, you will most likely continue to use regular computers at home.
Instead, we know of specific (but important!) applications for which a quantum computer promises a speed-up. In this part of the magazine we present a selection of different applications proposed in the scientific literature, ranging from algorithms that will need a universal and error-corrected quantum computer to applications designed for and running on smaller, noisy quantum processors.
A quantum processor beats classical computers
In 2019, scientists at Google reported using a 53-qubit processor to sample the output of a “random” quantum circuit. By measuring the resulting strings of 0s and 1s, they verified that the output corresponded to random numbers following a quantum probability distribution that is very hard to calculate on a classical computer. On their quantum processor, this task required minutes but may have taken thousands of years on a normal computer with the state of the art of 2019. Since then, more research has also improved the simulations on classical supercomputers, and this has blurred the line as to whether or not their results were truly faster. Nonetheless, this experiment illustrates the ability for quantum processors to out-muscle classical computers.
It’s important to ask: is this useful? While there are important applications for random number generation, it’s certainly not world-changing. It does show that modern quantum computing sits in a regime near that of the world’s best supercomputers, but still shy of the most modest useful quantum algorithms. To span this gap will require both technological advances and algorithmic ingenuity.
