A team of researchers led by physicist Tigran Sedrakian from the American company BlueQubit claims to have demonstrated a quantum advantage using an IBM processor. The quantum computer completed one million results in 19 seconds. The researchers estimate that a classical computer, such as Frontier, one of the most powerful supercomputers in the world, would require approximately 110 years to perform a comparable operation. Classical computers operate with bits, which are either 0 or 1, while quantum computers use qubits, which can exist in a combination of 0 and 1 before measurement. This allows quantum computers to approach certain problems in a radically different way. To test their performance, researchers use an experiment called Random Circuit Sampling. The principle involves the quantum computer performing a long series of randomly chosen operations, and then examining the results. The figure of 110 years does not correspond to the time required for a classical calculation in the general sense but is an estimate of the cost of classical simulation of this specific experiment. The researchers used a method called tensor network contraction to estimate the resources needed to reproduce the results on a classical machine. According to their calculations, reproducing the one million samples would require approximately 1.2 × 10²⁷ operations. By comparing this amount to the sustained performance of the supercomputer Frontier, they estimate the simulation would take about 110 years. However, this estimate must be considered with caution, as it depends on the simulation method and the performance of the classical machine used as a reference. New algorithms could significantly reduce the time required for a classical simulation. This experiment illustrates what the researchers call a "quantum advantage": in a task designed to exploit the properties of quantum computers, the quantum machine can produce a result much more quickly than a classical simulation. This type of demonstration is not new, as Google had previously announced a Random Circuit Sampling experiment with its Sycamore processor in 2019. The particularity of the new result lies mainly in the fact that the experiment was conducted on a commercial quantum processor accessible via IBM's cloud. The researchers claim to have used the standard environment of the service without any customized tuning of the processor. They have also published their circuits, results, and analysis code so that other users can try to reproduce the experiment.