Google Quantum AI says it has reached a new milestone in quantum error correction using a superconducting processor with roughly 1,000 qubits. The company said it was able to demonstrate a logical qubit with error rates below the level of the physical qubits used to build it, an important step toward making quantum computers more reliable.

The result does not mean practical quantum machines are ready for broad use, but it does indicate progress on one of the field’s biggest technical hurdles: keeping delicate quantum states stable long enough to do useful work. Error correction is widely seen as essential for scaling quantum systems beyond laboratory experiments and into tools capable of solving complex problems.

Google’s announcement adds to a fast-moving race among major technology firms and research labs to improve quantum hardware. While the company has not claimed that the new system is commercially viable, the achievement strengthens the case that larger and more resilient quantum processors are becoming technically possible.

Researchers and investors have long viewed error-corrected qubits as a key benchmark for quantum computing’s future. The latest update suggests the field is edging closer to that goal, though significant engineering challenges remain before quantum computers can reliably outperform classical systems on practical applications.