Xanadu, a Canadian quantum computing company, just announced it is targeting more than 1,000 'logical qubits' by 2031. That sentence contains two words that need unpacking before you can decide whether to care.
What a Logical Qubit Actually Is
A qubit is the quantum equivalent of a bit, the basic unit of information in a regular computer. But where a regular bit is either 0 or 1, a qubit can be in a superposition of both states at once, which lets quantum computers explore many solutions simultaneously. The catch: qubits are extraordinarily fragile. Any interference from the environment, heat, vibration, even stray electromagnetic fields, causes errors. A 'logical qubit' is a more stable, error-corrected version, built from many raw physical qubits working together to catch and fix each other's mistakes. Right now, making one reliable logical qubit typically requires hundreds of physical qubits. Nobody has done it at scale. Xanadu's announcement is a roadmap target, not a delivered product. It comes from a press release, not a peer-reviewed result.
Why Photons and Why This Approach Is Different
Xanadu uses photons, particles of light, as its qubits rather than the superconducting circuits that most competitors, including Anyon Systems, rely on. Photons do not need to be cooled to near absolute zero, which is one reason photonic approaches are appealing. The tradeoff is that photons are harder to make interact with each other, which is necessary for computation. Meanwhile, new research into hollow-core fiber is tackling a related bottleneck: how to connect different quantum devices to each other at all. Right now, quantum machines tend to be isolated systems; getting them to talk reliably is itself an unsolved engineering problem.
When Will Any of This Be Useful
Honest answer: not before 2030, and probably later. The 1,000-logical-qubit threshold is significant because below it, quantum computers cannot outperform classical supercomputers on the problems that actually matter, drug discovery, materials science, cryptography. A 2023 paper in Nature by Bluvstein et al. demonstrated meaningful error correction on 48 logical qubits, which gives a sense of how far 1,000 still is. What 2031 would represent, if Xanadu delivers, is the beginning of genuine utility, not the end of the race. The companies and governments betting on quantum right now are buying options on a technology that will matter enormously, just not yet. Atom Computing just hired a CFO, which is what companies do when they are preparing to raise serious money or go public. The institutional machinery is moving even while the physics catches up.