reported

Reported evidence

Google Quantum AI and collaborators reported below-threshold surface-code memories on the Willow superconducting processor, including distance-5 and distance-7 codes. The manuscript first appeared on August 24, 2024; the Nature publication appeared online in December 2024.

Sources: Google Quantum AI and collaborators, research manuscript

interpretation

Why it matters

Suppressing logical memory errors as code size grows is evidence for a key error-correction requirement. It does not demonstrate a profitable application, unrestricted fault-tolerant computation or independent replication.

Sources: Google Quantum AI and collaborators, research manuscript

interpretation

Comparison boundary

Read the code distances, decoding conditions and error definition in the paper. Do not compare this memory result directly with a vendor's raw physical-qubit total. Commercial usefulness is not established by this example.

Sources: Google Quantum AI and collaborators, research manuscript

reported

Source revision note

Source update reviewed October 10, 2026: the authors corrected Figure 3a’s repetition-code axis and legend labels on April 28, 2026. The notice concerns that repetition-code figure. It does not state that the main surface-code memory result was retracted. This guide does not reproduce numerical benchmark claims.

Sources: Nature

Sources and review scope

Source-linked. Editorial review: .

  1. Quantum error correction below the surface code threshold ↗

    Google Quantum AI and collaborators, research manuscript · Accessed 2026-10-02

  2. Author Correction: Quantum error correction below the surface code threshold ↗

    Nature · Accessed 2026-10-10 · Published 2026-04-28; full text reviewed 2026-10-10; Figure 3a axis and legend labels corrected

Keep exploring

Companies & Investing · Investing

Alphabet / Google Quantum AI

A quantum research program within a much larger advertising and cloud business.