Google claims quantum leap with new Willow chip

The breakthrough potential of quantum computers remains a ways off due to two crucial issues: error correction and computing power. In addition, the quality of qubits — aka quantum bits, the basic units of information for quantum computing — is not yet sufficient for longer calculations.

According to Hartmut Neven, founder and head of Google’s Quantum AI Lab, Google has achieved two decisive improvements with its new quantum chip Willow:

• First, Willow can reduce errors exponentially when scaling with more qubits, thus overcoming a central challenge in quantum error correction that has been worked on for almost 30 years.
• Second, Willow can perform a standard benchmark calculation in less than five minutes. One of the fastest supercomputers currently in existence would need ten quadrillion (10 to the power of 25) years to do the same — a number that far exceeds the age of the universe.

Better error correction

To calculate performance, Google used the Random Circuit Sampling (RCS) benchmark. RCS was developed in-house by the Quantum AI Lab and is widely used in the field of quantum computing.

Almost more important than the successes in terms of computing power are the advances in error correction. Until now, the fatal fact was that the number of qubits used generally increased along with the frequency of errors.

Google now wants to break this vicious circle. “The more qubits we use in Willow, the more we reduce errors and the more quantum the system becomes,” Neven claims.

To do this, Google’s team tested increasingly larger arrays of physical qubits, scaling up from a grid of 3×3 encoded qubits to a grid of 5×5 and then to a grid of 7×7. The result: Each time, the Google team was able to halve the error rate using its latest advances in quantum error correction. In other words, an exponential reduction in the error rate was achieved.

Commercial apps are getting closer

At the same time, the researchers made further progress: They were able to improve the quality of their qubit arrays, as these now have a significantly longer lifespan than the individual physical qubits. In other words, calculations can take longer.

For Neven, this has resulted in the most convincing prototype for a scalable logical qubit to date. He sees this as a sign that useful, very large quantum computers can actually be built. Willow thus brings the implementation of practical, commercially relevant algorithms that cannot be replicated on conventional computers, Neven claims.

Earlier this year, Microsoft reported a quantum breakthrough with a qubit-virtualization system that it claimed broke a logical-qubit creation record. The company also aims to provide a commercial offering.

Recent breakthroughs, including Chinese researchers’ work to break RSA encryption with a quantum computer, have advisory bodies urging CIOs and CISOs to prepare post-quantum encryption resiliency plans.