In a major leap forward for quantum computing, physicists at the University of Oxford have achieved the most accurate control of a quantum bit, or qubit, ever recorded.
The team reached an astonishingly low error rate of just 0.000015%—that’s only one mistake in every 6.7 million operations.
To put that into perspective, you’re more likely to be struck by lightning in a year than for one of their qubit operations to go wrong.
The research, soon to be published in Physical Review Letters, marks a critical step in building powerful quantum computers that can solve real-world problems.
Professor David Lucas, a co-author from Oxford’s Department of Physics, said the result sets a new global benchmark and could dramatically reduce the cost and complexity of future quantum machines.
Quantum computers rely on qubits to process information in ways far beyond the capabilities of traditional computers.
But for them to work effectively, the operations—or logic gates—that manipulate qubits must be extremely precise. If too many errors occur, the final result becomes useless.
While error correction can help fix some of these issues, it requires extra hardware and many more qubits, which makes the computers larger and more expensive.
That’s why this record-breaking achievement is so important.
Reducing the error rate means fewer extra qubits are needed to correct mistakes, making quantum computers smaller, faster, and more efficient.
Molly Smith, a graduate student and co-lead author, explained that this level of control not only improves quantum computing but can also benefit other technologies like quantum clocks and sensors.
The team achieved this accuracy using a single calcium ion as their qubit, chosen for its natural stability and long-lasting quantum properties. Instead of using the usual method involving lasers to control the ion, the Oxford researchers used microwave signals—more stable, cost-effective, and easier to integrate into quantum hardware.
Impressively, they ran their experiment at room temperature and without the need for special magnetic shielding. That’s a big deal, as it lowers the technical demands of building and running a quantum computer.
Oxford had already held the previous record in 2014 with an error rate of one in a million. Their continued leadership in the field helped launch Oxford Ionics in 2019, a spinout company that has become a leader in advanced trapped-ion quantum platforms.
Still, the team notes there’s more work ahead. While single-qubit operations are now incredibly accurate, two-qubit operations—where qubits interact with each other—are still much more prone to errors. The best results so far show one error in every 2,000 two-qubit operations, so improving those will be key to creating fully functional, fault-tolerant quantum computers.
This work was done in Oxford’s Department of Physics by Smith, Aaron Leu, Dr. Mario Gely, and Professor Lucas, along with visiting researcher Dr. Koichiro Miyanishi from Osaka University.
The team is part of the UK Quantum Computing and Simulation Hub, a national effort to advance quantum technologies.
