Google has announced a significant advancement in correcting errors inherent in today’s quantum computers, a crucial step toward overcoming the most challenging technical barrier in developing this revolutionary technology. The findings were published in the journal Nature.
Quantum computers face difficulties in producing useful results because qubits, the fundamental units of quantum information, maintain their quantum states for only a fraction of a second. This fleeting stability results in information loss before calculations can be completed. Addressing these errors is the primary technical challenge in the industry.
While some quantum startups focus on programming today’s error-prone, or “noisy,” machines for marginal improvements over traditional computers, these efforts have yet to yield practical results. The consensus is growing that quantum computing will only become useful once the error correction problem is resolved.
Google’s researchers have developed a method to distribute information across multiple qubits, allowing the system to retain enough information to complete calculations despite individual qubits losing their quantum states. Their research demonstrated a 4 percent reduction in the error rate as they scaled up their technique to a larger quantum system. Importantly, this marks the first instance where increasing the system size did not result in a higher error rate. This achievement shows Google has reached a “break-even point,” paving the way for continuous performance improvements and progress toward a practical quantum computer.
The breakthrough was achieved through enhancements in all components of Google’s quantum computer, including the quality of qubits, control software, and cryogenic equipment used to maintain near-absolute zero temperatures.
Google described this breakthrough as only the second of six steps necessary to develop a practical quantum computer. The next step involves refining their engineering to require only 1,000 qubits to create a “logical qubit”—an error-free abstraction built on top of imperfect physical qubits.
For more details see: Suppressing quantum errors by scaling a surface code logical qubit.
