Quantum Supremacy
Quantum supremacy (also called quantum advantage) is the milestone at which a programmable quantum computer performs a specific task that no practical classical computer could finish in a reasonable time. In 2019 a team at Google reported reaching this milestone with a 53-qubit processor named Sycamore.
The Sycamore experiment
Sycamore used 53 superconducting qubits cooled to near absolute zero. The chosen task was random circuit sampling: apply a long, random sequence of quantum gates and then measure the qubits many times. Because of quantum interference, the possible 53-bit outcomes are not equally likely — the circuit produces a characteristic, lumpy probability distribution. Sampling from that distribution is easy for the quantum hardware but extremely hard to simulate classically, because the number of amplitudes a classical computer must track grows as 253, roughly 1016.
Google reported that Sycamore drew a million samples in about 200 seconds, and estimated that the leading supercomputer of the time would need on the order of 10,000 years to reproduce the result. IBM contested that estimate, arguing that a better classical algorithm using disk storage could finish in a few days — a useful reminder that the supremacy boundary is a moving target as classical methods improve.
What it does and does not prove
The experiment demonstrated that a quantum device can outpace classical computation on a carefully chosen benchmark. It did not solve a practically useful problem — random circuit sampling has no direct application — nor did it deliver an error-corrected, general-purpose quantum computer. Its real significance is as a proof of principle: large, controllable quantum systems can reach a regime that is genuinely beyond classical reach.
A common misconception
Quantum supremacy does not mean quantum computers are now better than classical computers at everyday tasks. They remain worse at almost everything. The claim is narrow and deliberate: one contrived problem, run faster than classical simulation, to mark a scientific threshold rather than a commercial one.