The Quantum Investor's Notebook logo The Quantum Investor's Notebook
← Back

Monday, June 15, 2026

The Logical Qubit: Safety in Numbers

A single physical qubitAn actual piece of qubit hardware. On its own it is fragile and makes frequent errors. is too fragile to trust. The logical qubitA reliable 'qubit' built by bundling many error-prone physical qubits together with error correction. These are the units that actually matter for useful computing. is the trick that turns a crowd of unreliable qubitsThe basic unit of a quantum computer. Like a 'bit' in a normal computer, but instead of being only 0 or 1 it can be 0, 1, or a blend of both at once. into one you can actually compute with.

What a logical qubit actually is

The thing a logical qubit is, is one unit of quantum information spread across many physical qubits at once, so that no single one of them holds the answer.1 A lone physical qubit, a superconducting circuit or a trapped ionA qubit made from a single electrically charged atom held in place by electromagnetic fields and controlled with lasers., is hopelessly delicate; the faintest nudge from its surroundings corrupts it, a process called decoherenceThe loss of a qubit's quantum information as it interacts with the outside world. The main reason qubits are fragile.. The logical qubit's bet is that a crowd can be steadier than any of its members. You do not protect the information by building one perfect qubit. You spread it out, and you watch the crowd for trouble.

You do not protect the information by building one perfect qubit. You spread it out, and watch the crowd for trouble.

That watching is the clever part. You can check a quantum system for errors without reading the information it carries, which would destroy it. Extra checks, called syndrome measurementsExtra measurements that reveal whether and where an error happened, without reading the protected quantum information itself., ask only whether something just went wrong and where, never what the value is.2 That is the heart of quantum error correctionTechniques that combine many shaky physical qubits into fewer reliable ones, so a long calculation stays correct.: catch the slips as they happen, undo them, and the encoded qubit outlives every one of its fragile parts.

On the left, a single physical qubit drawn as one shaky circle, fragile on its own. An arrow labeled spread out and check for errors points to the right, where many physical qubits form a grid grouped as one logical qubit, and an error on one qubit is caught and fixed. Many fragile parts make one reliable whole.
Figure 1: A single physical qubit is fragile on its own. Spread one unit of information across a grid of them and watch for errors, and the crowd becomes one reliable logical qubit, even as individual members slip.

How it is built, and what it costs

The leading way to do this is the surface codeThe leading error-correction scheme for superconducting qubits. It needs many physical qubits to protect one logical qubit.: physical qubitsAn actual piece of qubit hardware. On its own it is fragile and makes frequent errors. laid out on a flat, two-dimensional grid where each one only ever talks to its neighbours, which is about all real hardware can manage.3 The dial that matters is the code distanceThe size of an error-correcting code: the fewest physical errors that can corrupt the logical qubit. Bigger distance means stronger protection., essentially the size of the grid. Here is the fact that makes the whole enterprise possible: as long as the hardware's error rate sits below a critical threshold, making the grid bigger pushes the logical error rate down, and it falls fast.3

Below threshold, more qubits means a better qubit. Above it, more qubits just means more ways to fail.

None of this is cheap. Estimates for a serious calculation run to thousands, even tens of thousands, of physical qubits for every single logical one.3 That overhead is the central price of quantum computing, and it is the honest reason a raw qubit countThe raw number of physical qubits in a machine. A weak measure of power on its own, since quality and connectivity matter more. tells you so little.

Break-even, and the real finish line

For years this was all on paper. The milestone everyone waited for was break-even: the point where a logical qubitA reliable 'qubit' built by bundling many error-prone physical qubits together with error correction. These are the units that actually matter for useful computing. finally outlives its best single physical qubit. Recent experiments have crossed it, running a logical qubitThe basic unit of a quantum computer. Like a 'bit' in a normal computer, but instead of being only 0 or 1 it can be 0, 1, or a blend of both at once. that beats its own parts.4 That is a genuine turning point. But it is only the first one. The harder proof, the one that actually matters, is that the logical qubit keeps getting better as the grid grows.

I will be honest about what a logical qubit is and is not. It is not a faster qubit, and it is not the finish line. It is the unit a machine built for fault toleranceThe milestone where a quantum computer can run long calculations correctly despite ongoing errors. It is the field's holy grail. is assembled from, the thing that lets a long calculation run to the end without decoherenceThe loss of a qubit's quantum information as it interacts with the outside world. The main reason qubits are fragile. erasing the answer before you get there. Strip it away, and large-scale quantum computing simply does not happen.

Sources

  1. Devitt, S. J., Munro, W. J. & Nemoto, K. “Quantum Error Correction for Beginners.” Reports on Progress in Physics 76, 076001 (2013). DOI: 10.1088/0034-4885/76/7/076001. Preprint: arXiv:0905.2794.
  2. Terhal, B. M. “Quantum Error Correction for Quantum Memories.” Reviews of Modern Physics 87, 307–346 (2015). DOI: 10.1103/RevModPhys.87.307.
  3. Fowler, A. G., Mariantoni, M., Martinis, J. M. & Cleland, A. N. “Surface Codes: Towards Practical Large-Scale Quantum Computation.” Physical Review A 86, 032324 (2012). DOI: 10.1103/PhysRevA.86.032324. Preprint: arXiv:1208.0928.
  4. Google Quantum AI. “Quantum Error Correction Below the Surface Code Threshold.” Nature 638, 920–926 (2025). DOI: 10.1038/s41586-024-08449-y. Preprint: arXiv:2408.13687.