What was shown
The team ran this on a physical quantum processor. They stored one quantum bit of information as a pattern spread across the vibrations of a microwave cavity. A single helper qubit checked that pattern again and again. A fast classical controller then applied a correction after each check. Think of a sandcastle that gets small repairs before every wave, so the damage never builds up. The test was simple: does the repaired version last longer than the best version left alone? It did. The corrected bit held its information for longer than every imperfect component used to build it. The paper calls this gain beyond break-even, meaning correction finally helps more than it costs. The corrected bit still makes small errors in each correction round. They are small enough that the net effect is a clear improvement.
“We demonstrate full code stabilization and error correction with gain G=2.27±0.07 using the Gottesman-Kitaev-Preskill (GKP) encoding [2] of a logical qubit into grid states of an oscillator.”
— Introduction (main text, before Fig. 1); confirmed in section 'Observing QEC beyond break-even', in: Real-time quantum error correction beyond break-even1
- How it was done: run on a physical quantum processor
- Qubits: One protected quantum bit, stored in a microwave cavity and checked by a single helper qubit.
- Error correction: Active correction ran in real time on existing hardware, with a classical controller applying fixes after each check.
- Compared against classical computing: no comparison stated in the paper
Who it affects
This concerns companies building quantum computers and the customers who wait for machines that can run long calculations reliably. The result is a storage step, not an application. For a business reader, nothing changes in what can be bought or used today. What changes is the evidence: one route to error-corrected quantum hardware now has a measured case where correction pays off instead of only costing effort.
Quantum computing vendors and their customers: McKinsey & Company estimates $43B to $71B by 2035 — quantum computing market size (vendor side) (Quantum Technology Monitor, April 2026, p. 8).
This figure is an estimate, not a measurement: McKinsey & Company is a consultancy with its own quantum advisory business. It covers a whole use case, not the result of this paper.
What is still missing
The result is one protected qubit used as a memory. The paper does not demonstrate operations on that qubit, several protected qubits working together, or a larger error-correcting scheme built on top of such qubits. The authors also report occasional collapses of performance tied to stray, unwanted degrees of freedom in the system. They also report sensitivity to errors in the helper qubit, and both remain to be solved. Until these are fixed and operations are added, this memory cannot yet run a calculation.
Who is working on it
The paper (November 2022) lists its authors' affiliations as: Departments of Physics and Applied Physics, Yale University, New Haven, CT 06520, USA; Yale Quantum Institute, Yale University, New Haven, CT 06511, USA; Institut Quantique and Département de Physique, Université de Sherbrooke, Québec, J1K 2R1, Canada. That is who signed this work — it is not a ranking of who is ahead.
This strengthens error correction that stores quantum information in the vibrations of superconducting cavities, checked by a helper qubit. It also strengthens the use of reinforcement learning, a trial-and-error method that needs no model of the hardware, to tune the correction circuits while the experiment runs.
How this brief was made
This brief was written by software, not by a person: a language model read the full text of the paper, and code then checked every quotation, number and name in this brief against that text before it was published. The rules are set out under Method. Responsible for the rules and for this page: Georg Singer. The paper is a preprint that has not been peer-reviewed. Georg Singer holds no shares in any company or institution named in this brief (as of October 2026).