Untangling the challenges of quantum computing - Nature

Download PDF From scaling to error correction, numerous problems remain to be addressed before a practical quantum computer can be built, but researchers continue to find the necessary solutions. Quantum computers could be used to solve complex problems that would be unfeasible for the most powerful conventional computer. But building a practical quantum computer is, of course, far from easy. To start, you need a scalable qubit platform. You need to be able to control and readout the state of the qubits. You need to be able to entangle multiple qubits. And you need to be able to determine if errors have occurred and correct them. Each of these tasks comes with its own particular challenges. Nevertheless, researchers in academia and industry continue to find solutions, bringing useful machines ever closer to reality1. In this issue of Nature Electronics, we highlight some of the latest advances.Optical microscopy image of a section of the single-flux quantum control carrier chip used in the quantum processor unit developed by Shu-Jen Han and colleagues. Credit: SEEQC Inc.Superconducting qubits are currently one of the leading platforms in the race to build a practical quantum computer. Such a machine will, however, potentially require a million or more physical qubits2,3, and each qubit typically uses an individual signal line to control it. This makes scaling challenging — more qubits mean more wires. The platforms also often have limited integration as the qubits operate at millikelvin temperatures and the control electronics at room temperature. In an Article in this issue, Shu-Jen Han and colleagues report a quantum processor unit in which superconducting qubits and single-flux quantum control electronics are integrated into a single multi-chip module via flip-chip bonding. With the approach, the qubit layer and the control electronics can operate on the same millikelvin temperature stage. The researchers — who are based at Seeqc Inc. in New York and Seeqc UK in London — also use digital demultiplexing to distribute control pulses to multiple qubits and, with this, break the linear scaling of control lines to number of qubits.Distributed quantum networks— in which multiple quantum chips are linked together via communication channels — can help in the scaling of quantum computing. And microwave technology, which is the basis of current telecommunications, could assist in the creation of such networks. But microwave photons are sensitive to thermal noise, which typically destroys quantum state information during transit. In a further Article this month, Jingjing Niu, Youpeng Zhong, Dapeng Yu and colleagues — who are based at the International Quantum Academy in Shenzhen, Southern University of Science and Technology in Shenzhen, Ningxia University, and Hefei National Laboratory — report a thermal-noise-resilient microwave quantum network. The approach decouples the millikelvin qubit operating temperature from that of the communication channel, operating at 4 K.

As Peter Rabl of the Technical University of Munich notes in an accompanying News & Views article, “The protocol is also not restricted to 4 K, and could potentially be integrated with high-temperature superconductors operating at liquid-nitrogen temperatures of 77 K.”Silicon spin qubits could exploit the manufacturing techniques used to build conventional computers and thus offer a promising route to scalable quantum processors. The readout of such qubits typically involves radiofrequency single-electron transistors. But these suffer from a trade-off between chip space and fidelity. In another Article in this issue, Jacob Chittock-Wood, John Morton, Fernando Gonzalez-Zalba and colleagues report a radiofrequency electron cascade readout method for coupled spin qubits.

The team — who are based at University College London, Quantum Motion, the University of Cambridge, the University of Oxford and imec — use the approach to perform singlet–triplet readout of two electron spins in a natural silicon planar metal–oxide–semiconductor quantum dot array.The large-scale integration of spin qubits will also require approaches to characterize quantum components at scale. To address this, Giordano Scappucci and colleagues at Delft University of Technology report a crossbar chip for benchmarking semiconductor spin qubits. A further challenge in the scaling of these semiconductor spin qubits is the delicate tuning required to reach and maintain qubit operation. To address this, Jonas Schuff, Natalia Ares and colleagues at the University of Oxford, University of Basel and Mind Foundry in Oxford report a fully autonomous tuning process for spin qubits.Practical quantum computation probably requires qubit error rates lower than 10−10 (ref. 4). Errors are though inevitable in physical qubits, and the error rates of spin qubits fall short of this threshold. Error detection and correction is thus essential. In a further Article this month, Guanyong Wang, Guangchong Hu, Yu He, Dapeng Yu and colleagues report quantum error detection in a silicon quantum processor. The researchers — who are based at Southern University of Science and Technology in Shenzhen, the International Quantum Academy in Shenzhen, and Hefei National Laboratory — use a system composed of four nuclear spin qubits and one electron spin qubit, and show that an arbitrary single-qubit error can be detected. (See also the accompanying News & Views article about the work from Lieven Vandersypen of Delft University of Technology.) ReferencesCastelvecchi, D. Nature 650, 24–26 (2026).Article Google Scholar Potočnik, A. Nat. Electron. 8, 3–4 (2025).Article Google Scholar Megrant, A. & Chen, Y. Nat. Electron. 8, 549–551 (2025).Article Google Scholar Acharya, R. et al. Nature 638, 920–926 (2024).

Google Scholar Download referencesRights and permissionsReprints and permissionsAbout this articleCite this article Untangling the challenges of quantum computing. Nat Electron 9, 237 (2026). https://doi.org/10.1038/s41928-026-01607-2Download citationPublished: 30 March 2026Version of record: 30 March 2026Issue date: March 2026DOI: https://doi.org/10.1038/s41928-026-01607-2Share this articleAnyone you share the following link with will be able to read this content:Get shareable linkSorry, a shareable link is not currently available for this article.