Quantum History Part 5: The Error Correction Era (2023 to 2026)
The turn: quality over quantity
By 2023 the field had learned that adding noisy qubits does not add up to a useful machine. The new goal was the logical qubit: a reliable qubit made from many physical ones using error correction. The race since then has been about showing that bigger codes make errors smaller, not larger. This is the most exciting stretch of the whole story, because it is the road to machines that can run long programs.
April 2024: four logical qubits
Quantinuum announced on April 3, 2024 that, working with Microsoft's error correction techniques, it built four logical qubits from 30 of the 32 physical qubits on its H2 processor. It reported logical error rates about 800 times lower than the corresponding physical error rates. That is a company claim, but it fit the broader direction of travel.
December 2024: Google's Willow
On December 9, 2024, Google announced Willow, a 105 qubit chip. Its team tested encoded arrays of 3 by 3, 5 by 5 and 7 by 7 qubits, and reported the logical error rate roughly halved at each step. Google calls this below threshold, the long sought sign that you can scale up and win. The results were published in Nature. The same announcement claimed Willow finished a random circuit sampling task in under five minutes that Google estimates would take Frontier about 10 septillion years, using what it says were generous assumptions; that benchmark has no known commercial use. More in Google Quantum AI and The Surface Code Explained for Beginners.
2025: a crowded year
February. Microsoft announced Majorana 1 and a path to topological qubits. Physicists were mostly unconvinced by the public evidence, and Nature's editors noted the paper did not show evidence of topological modes. Treat it as an open question; see Topological Qubits Explained.
May. Craig Gidney of Google posted a preprint estimating RSA-2048 could be factored in under a week with fewer than a million noisy qubits, down from a 2019 estimate of about 20 million. It is a resource estimate, not an attack, and no existing machine comes close.
October. On October 22, Google described Quantum Echoes on Willow, which it says ran 13,000 times faster than the best classical algorithm on a leading supercomputer, and is "quantum verifiable". A related molecule demonstration was a proof of principle, matching standard NMR results.
November. IBM unveiled the 120 qubit Nighthawk and an experimental chip called Loon, with a roadmap, reported in press coverage, toward a fault tolerant machine called Starling by 2029 (see IBM's Quantum Roadmap). Quantinuum launched Helios with 98 physical qubits and claimed 48 error-corrected logical qubits at a 2 to 1 encoding rate, plus 94 error-detected logical qubits. One of its scientists said publicly the two headline numbers were not a single combined result, which is a good reminder to read the fine print. See Quantinuum Ltd.
2026 so far
- February: Iceberg Quantum, a Sydney startup, reported a design that might factor RSA-2048 with under 100,000 physical qubits, based on simulation. Reported, not demonstrated.
- March 31: Google Quantum AI and collaborators reported that the 256 bit elliptic curve problem might fall to fewer than 500,000 physical qubits, down from about 9 million in a 2023 estimate. The details were held back, with a zero knowledge proof released instead. Read the implications in Will Quantum Computers Break Bitcoin and Solana.
- September 24: Infleqtion reported 30 entangled logical qubits from 80 physical qubits on a neutral atom machine. This is a company press release, using error detection, and has not been independently verified.
- November 2025: IBM was among the 11 companies DARPA moved into Stage B of its Quantum Benchmarking Initiative, an independent look at routes to fault tolerance; see DARPA's Quantum Benchmarking Initiative.
Why to be optimistic, and careful
The trend is clear: error rates are falling as codes grow, more companies are showing logical qubits, and independent programs are testing the claims. The cautions are also clear. Logical qubit counts are not comparable across companies, many results are error detection rather than full correction, and useful applications are still ahead. The state of play is summarized in Quantum Error Correction in 2026. The final chapter covers the biggest debates.
This page is education, not financial advice. The QNT memecoin is independent and unrelated to Quantinuum Ltd, Google, IBM or any lab or government.
Sources and further reading
- Quantinuum: four reliable logical qubits with Microsoft
- Google: Meet Willow
- Google: Quantum Echoes
- Google Research: disclosing quantum vulnerabilities responsibly (March 2026)
- The Quantum Insider: Helios reception
- Science News: physicists mostly unconvinced by Microsoft's chip
- The Quantum Insider: three papers on the quantum threat timeline
- Infleqtion: 30 entangled logical qubits
- IBM: Stage C of DARPA QBI
Reported as of 2026-10-09. Company claims are the companies' own unless a source says otherwise. Educational content only, not financial advice. The QNT memecoin is independent of Quantinuum Ltd and every other company, lab or government.
Frequently asked questions
What is a logical qubit?
A reliable qubit made by encoding information across many physical qubits so errors can be detected and corrected. Counts of logical qubits are not directly comparable between companies.
What did Google's Willow show?
Google reported that larger encoded arrays (3 by 3, 5 by 5, 7 by 7) halved the logical error rate at each step, which it calls below threshold performance. The work was published in Nature.
Can quantum computers break RSA today?
No. Gidney's 2025 estimate of under a million noisy qubits is a theoretical estimate, and current machines are far smaller and less reliable than that.
Keep reading
- Quantum Error Correction in 2026: Where the Race Really Stands
A plain English scorecard of error correction progress: below-threshold results, logical qubit demonstrations, magic states, qLDPC codes and what is still unproven. - Google Quantum AI: Willow, Error Correction and Quantum Echoes
What Google's Willow chip and the Quantum Echoes result showed, why below-threshold error correction matters, and what is still unproven. - Quantinuum Ltd: Helios, the IPO and Why the Name Overlaps with QNT
A neutral profile of the quantum computing company Quantinuum Ltd, its Helios computer and its reported 2026 Nasdaq listing, and how it differs from the QNT memecoin. - Quantum Computing's Biggest Debates: Hype, Skeptics and Real Progress
Supremacy versus advantage, classical rivals, the noise skeptics, dequantization and the Majorana dispute, told fairly.
All Quantum computing guides | Back to top | Search the site
Main pages: Quantum computing explained | Quantum and crypto | Companies | Quantum news | Glossary