What the experiment actually showed · Who did it, and where the first-party record is · Why this is a research column, not a financing event · Context: this is the second AQEC break-even result in weeks
What the experiment actually showed
The team encoded a photonic logical qubit in a three-dimensional superconducting microwave cavity using the lowest-order binomial code. Its dominant error is single-photon loss. A quantum-optimised control pulse maps the error state coherently back into the code space while moving the error entropy to an auxiliary transmon; a Rabi-driven dissipation channel then resets that ancilla. The loop needs neither syndrome measurement nor real-time classical feedback — the defining feature of autonomous quantum error correction (AQEC).
Under repeated AQEC, the logical lifetime reached about 2.0 ms, 18% above the best physical qubit in the same system. The same correction was applied to a small frequency-shift measurement and produced a 6.3 dB metrological gain over an uncorrected Fock-state probe. Crossing break-even means the protected logical qubit survives longer than the best uncorrected physical qubit in that system; it is a system-relative benchmark, not a universal record and not a claim about logical gate fidelity.
Who did it, and where the first-party record is
The institute's release names Ni Zhongchu, Hu Ling and Cai Yan Yan as joint first authors, Zheng Shibiao of Fuzhou University and Xu Yuan of Shenzhen International Quantum Academy as joint corresponding authors, with Yu Dapeng among the participants. The paper, 'Extending qubit lifetime beyond break-even and enhancing quantum sensing by autonomous quantum error correction', was published in Nature Physics on 16 November 2026 (Nature Physics paper).
The WeChat article summarises that release accurately but compresses the method and omits the quantitative gains. This site therefore cites the institute announcement and the journal paper as the sources of record and treats the WeChat piece as a secondary pointer to them — the same hierarchy used for company claims: a first-party document outranks a repost, however well written.
Why this is a research column, not a financing event
QuantumRoundup tracks capital: disclosed rounds, public-market instruments and government programmes, each with a source and an editorial note. A peer-reviewed result is a different kind of fact. It has no amount, no investor, no company beneficiary and no closing date. Entering it as a funding row would invent a transaction the paper does not describe.
The institutions involved are research organisations, not companies in this site's registry, and none of the named researchers has a people-directory entry here. There is consequently no honest internal link to add: adding one would either point at a record that does not exist or imply a company affiliation the paper does not state. The data table below lists the open files this site does publish, so a reader can check exactly which claims are counted here.
Context: this is the second AQEC break-even result in weeks
On 23 October 2026 a University of Science and Technology of China team published a separate Nature Physics result, 'Beating the break-even point with autonomous quantum error correction', using a four-level code on a single trapped calcium ion and reporting a 42.7 ms fitted logical lifetime against 29.4 ms for the best physical qubit. Two platforms reaching break-even with autonomous, measurement-free correction within a month is a meaningful convergence — but neither result changes a funding number on this site.
What this site actually publishes as open data (computed on this build)
| File | What it contains |
|---|---|
| companies.json | Company registry, verified dossiers and cumulative disclosed totals |
| company-intel.json | Editor-verified China company records, exported from the registry |
| funding-events.json | Raw report ledger, including duplicates and candidates awaiting a note |
| stats.json | Aggregates used by the hero numbers and snapshots |
| rankings.json | De-duplicated, rendered view used by the tables |
The practical takeaway
A scientific breakthrough and a financing event are both news, but only one of them is money. This page keeps the Nature Physics result on the record with its first-party source and its measured gains, and leaves the funding ledger untouched — which is what makes the ledger usable when the next round is announced.