Google has introduced a new quantum chip called Willow. It performs calculations so fast that developers assume that they take place in parallel worlds.
The developers claim that they have managed to halve the number of quantum errors, which remain one of the main obstacles in the field of quantum computing.
Meet Willow: Our state-of-the-art quantum chip. It's the first quantum chip to show exponential error reduction as qubits scale, paving the way for large-scale, fault-tolerant quantum computers. Dive in → https://t.co/Lr1vkZk1QT pic.twitter.com/8VkiXQ694u
Meet Willow: Our state-of-the-art quantum chip. It's the first quantum chip to show exponential error reduction as qubits scale, paving the way for large-scale, fault-tolerant quantum computers. Dive in → https://t.co/Lr1vkZk1QT pic.twitter.com/8VkiXQ694u
— Google Quantum AI (@GoogleQuantumAI) December 9, 2024Quantum computers operate on the basis of qubits, units of information that can simultaneously be in states 0 and 1 due to the phenomenon of superposition. However, qubits are prone to rapid information exchange with the external environment, which makes it difficult to store the data required to complete calculations. Usually, an increase in the number of qubits leads to an increase in the number of errors.
But in Willow, the developers managed to achieve the opposite situation: the more qubits, the fewer errors.
Google evaluates quantum performance using the Random Circuit Sampling (RCS) benchmark. According to the developers, every team creating a quantum computer should first test its ability to outperform classical computers in this indicator.
The newest Willow chip performed the RCS calculation in less than 5 minutes, while a classical computer would have needed 10²⁵ years (10 septillion years) to do so. For comparison, this is much longer than the age of our Universe, which is about 14 billion years.
The developers report that this suggests that quantum computing takes place in many parallel universes and confirms the existence of a multiverse.
However, the researchers shared another important performance metric. They were able to increase the average qubit excitation time (T₁) by almost five times, from about 20 µs in the previous Sycamore chip to 68 µs ± 13 µs.
However, the RCS calculation has no practical application and only serves to measure the performance of quantum computers. So the next step for the developers is to demonstrate the first useful, non-classical calculations that are relevant to real-world applications.
