The demonstration of a quantum computational advantage is a milestone worth celebrating.
Last month, Nature published one of the worst kept secrets in physics: the experimental demonstration of ‘quantum supremacy’ reported by a collaboration led by the Google AI Quantum team.
The feat was achieved by Sycamore, a quantum processor made up of 53 superconducting circuits operating below 20 mK, which took 3 minutes to carry out a calculation that is estimated to take 10 millennia on our best classical supercomputers. It consists of sampling from the output distribution of a random quantum circuit, a task with essentially no other purpose than demonstrating a quantum computational advantage.
Google AI Quantum’s team had been pursuing this goal for some time. For many specialists in the field, it was a matter of when it would happen, rather than if. Nevertheless, the wider reaction to the report, first broken by the Financial Times in September after a preprint of the paper was accidentally posted on a public NASA server, was one of huge excitement. Pronouncements that we are at the dawn of a new technological era abounded, and for a short while many scientists felt uncomfortable, as often happens when a story gets too hot to handle.
More critical voices therefore quickly appeared to manage expectations, pointing out the relatively arbitrary nature of the threshold that constitutes quantum supremacy, and cautioning against the potential corporate interests behind a loud announcement with little practical substance. If quantum and classical machines are pitted against each other on a task that has been engineered to favour the former, the argument goes, should we really take this to be such a meaningful display of superiority? Outperforming classical machines on an actually meaningful task surely ought to be the real goal, and that’s likely still a long way off.
But there is little doubt that Google’s experiment represents an extraordinary feat of physics, computer science and engineering. If only as a good excuse to pause and celebrate the journey that took us from the dawn of quantum information science in the 1980s all the way to performing computations by navigating a Hilbert space of 253 dimensions, the achievement is well worth the title of a scientific milestone.
And as with all good milestones, quantum supremacy is really just the beginning. A few decades from now we may well acknowledge that this was, after all, a fairly arbitrary threshold. But it is one that seems destined to inspire a future generation of scientists.