Just now, Google CEO Pichai excitedly announced that Google's latest quantum computing chip, Willow, has been released! Willow achieved amazing results in benchmark tests: it completed a standard calculation in less than 5 minutes, which would take a top supercomputer more than 10^25 years - longer than the age of the universe!

When the news came out, even Musk exclaimed: "Wow!" 

Pichai also proposed using SpaceX's starship to build a quantum cluster in space in the future. Musk said: "This is likely to happen." Any self-respecting civilization should at least reach Kardashev Type II civilization. Musk believes that we are currently only less than 5% of Type I civilization. In order to reach about 30%, we need to place solar panels in all deserts or highly arid areas. 

Pichai agrees and says we should expand solar energy more instead of looking for alternatives

Sam Altman also congratulated

Willow's achievements are mainly reflected in two aspects

Exponential reduction of errors:

This has been a long-standing challenge in the field of quantum computing. Normally, the more qubits are used, the more errors will occur. But Willow has achieved an exponential decrease in the error rate as the number of qubits increases. The results published by Google in the top international journal Nature show that they tested qubit grids of different sizes from 3x3, 5x5 to 7x7, and each time they were able to halve the error rate. This is called "below threshold" and has been a difficult problem that has not been solved since Peter Shor proposed quantum error correction in 1995. This achievement is also one of the first successful cases of real-time error correction on superconducting quantum systems, which is crucial for any useful calculation, because if the errors cannot be corrected quickly enough, the calculation will be destroyed before it is completed. In addition, this is also a demonstration of "beyond break-even", in which the lifetime of the qubit array is longer than that of a single physical qubit, which is an unforgeable sign that error correction is improving the entire system. As the first system to go below the threshold, this is the most convincing prototype built to date for scalable logical qubits.

Performance beats supercomputers:

Google used the Random Circuit Sampling (RCS) benchmark to measure Willow's performance. Pioneered by the Google team and now widely used by the industry, RCS is the hardest classical benchmark that can currently be performed on a quantum computer. Willow's test results are astounding: it completed a calculation in less than 5 minutes that would take 10^25 years, or 10 trillion trillion years, on one of today's fastest supercomputers. This number far exceeds the known time scales in physics and far exceeds the age of the universe. This also confirms the idea that quantum computing occurs in multiple parallel universes, consistent with the idea that we live in a multiverse, a prediction originally proposed by David Deutsch. It’s worth noting that Google’s assessment of Frontier is based on conservative assumptions, such as full access to secondary storage (i.e., hard drives) and no bandwidth overhead—a generous and unrealistic allowance for Frontier

More technical details:

105 qubits: Willow has 105 qubits and achieves industry-best performance in two system benchmarks, quantum error correction and random circuit sampling

5x improvement in T1 time: T1 time (a measure of how long a qubit remains in an excited state) is close to 100 microseconds, a 5x improvement over the previous generation chip

Advanced manufacturing facility: Willow is produced in Google’s state-of-the-art manufacturing facility in Santa Barbara, one of the few in the world built from scratch for this purpose. Systems engineering is critical in the design and manufacturing of quantum chips: all components of the chip, such as single-qubit and two-qubit gates, qubit reset, and readout, must be well engineered and integrated simultaneously

Future Outlook, Advanced AI

Google’s next challenge is to demonstrate the first “useful, beyond-classical” computation on today’s quantum chips, which is relevant to real-world applications. They are optimistic that the Willow generation of chips can help them achieve this goal. They invite researchers, engineers, and developers to join them by checking out their open source software and educational resources (including their new course on Coursera), where developers can learn the essentials of quantum error correction and help them create algorithms that can solve future problems.

Pichai believes that

Advanced AI will benefit significantly from quantum computing. Quantum algorithms are supported by fundamental scaling laws, as we have seen with RCS. Similar scaling advantages exist for many basic computing tasks that are critical to artificial intelligence. Therefore, quantum computing will be indispensable for collecting training data that classical machines cannot access, training and optimizing certain learning architectures, and modeling systems where quantum effects are important. This includes helping us discover new drugs, designing more efficient batteries for electric vehicles, and accelerating progress in fusion and new energy alternatives. Many of these future game-changing applications will not be feasible on classical computers; they are waiting for quantum computing to unlock them

This is incredible. I personally feel that this is another achievement of Google that has changed human history after Transformer.