Physicists at the University of Oxford have achieved the most accurate control of a quantum bit (qubit) ever recorded, making just one mistake in 6.7 million single-qubit operations—an error rate of 0.000015 percent. This breakthrough, nearly ten times more precise than their previous world record set a decade ago, will be published in Physical Review Letters under the title Single-qubit gates with errors at the 10⁻⁷ level.
To illustrate how rare these errors now are, the team notes that a person is more likely to be struck by lightning in a given year (a probability of 1 in 1.2 million) than for one of their quantum logic gates to fail. This leap in reliability addresses one of the biggest obstacles to building practical quantum computers: maintaining accuracy across millions of operations.
Reliable quantum machines must perform millions of operations across large arrays of qubits. When error rates are too high, results become unusable. Error-correcting codes can preserve computation, but they require each logical qubit to be backed by clusters of physical qubits, dramatically increasing cost, size, and system complexity.
By sharply reducing the chance of error, the Oxford team’s work reduces the overhead required for error correction. This opens the door to future quantum computers that are smaller, faster, and more efficient. Precision in qubit control is also essential for advancing quantum technologies such as next-generation clocks and sensors, said Molly Smith, a graduate student in Oxford’s Department of Physics and co-lead author of the study.
The record-setting performance was achieved using a single trapped calcium ion, a qubit known for its long coherence time and inherent robustness. Instead of traditional laser pulses, the Oxford researchers controlled the ion’s quantum state using precisely tuned microwave signals. This method offers several practical advantages: it is cheaper, more stable, and easier to integrate with ion-trap chips than laser-based systems.
Crucially, the entire system operated at room temperature and without magnetic shielding, avoiding the need for complex cryogenic or isolation infrastructure. The team argues that these practical engineering gains are as important as the raw error rate when it comes to building scalable, deployable quantum processors.
The experiment was conducted by Molly Smith, Aaron Leu, Dr. Mario Gely, and Professor David Lucas, along with visiting researcher Dr. Koichiro Miyanishi from Osaka University’s Centre for Quantum Information and Quantum Biology. All are part of the UK Quantum Computing and Simulation Hub within the National Quantum Technologies Programme.
A fully functional quantum computer requires both single- and two-qubit gates operating together. While Oxford’s single-qubit gate error is now nearly one in seven million, the best two-qubit gates globally still produce errors roughly once every 2,000 operations. Reducing this figure to the same 10⁻⁷ level remains a critical milestone on the path to fault-tolerant quantum computing.
For now, the new single-qubit benchmark redefines what high fidelity means in quantum logic. It also provides a more realistic engineering roadmap: fewer qubits devoted to error correction, simpler electronics, and room-temperature systems that could move quantum computing out of research labs and into practical, real-world applications.
By Impact Lab
