28 Oct 2026

Advancing qubits

By Laurence Coldicott, Senior Content Director, Quantum.Tech
Advancing qubits

“How many qubits?” remains one of the first questions asked about a quantum computer. It is understandable: a rising number offers a simple way to describe progress in a complicated industry. 

But for the enterprise leaders deciding where quantum belongs in their technology strategy, the more useful question is: what can those qubits reliably achieve? 

Advancing qubits means improving their quality, connecting them effectively and keeping errors under control throughout a computation. The commercial significance will come from how these capabilities work together. 

Making reliability scale 

Quantum error correction sits at the heart of this challenge. By encoding information across multiple physical qubits, researchers can create a logical qubit designed to protect that information against errors. The crucial test is whether increasing the protection actually improves reliability. 

Google’s Willow research demonstration of a principle that practical, fault-tolerant quantum computing will depend on. It was also a quantum memory experiment, rather than evidence that commercially useful, fault-tolerant computation had arrived. That distinction matters when translating a scientific milestone into an enterprise expectation. [1] 

Protecting information is one step. Performing a long sequence of useful operations on it, while maintaining that protection, raises further demands. 

Research on neutral-atom processors has demonstrated logical operations and circuits involving up to 48 encoded logical qubits. These experiments broaden the evidence for computing with protected quantum information, although their results and error-handling methods need to be understood in context. A logical-qubit count alone does not provide a complete comparison between machines. [2] 

The engineering around the qubit 

The next stage also depends on what surrounds the processor. 

Control systems must operate accurately and consistently. Classical computing must interpret error measurements quickly enough to support correction. Connections between components must preserve quantum information. These requirements turn qubit development into a wider systems-engineering challenge. 

IBM’s work illustrates that relationship. Its fault-tolerance programme combines processor design, error-correcting codes, decoding and infrastructure. In August, the company announced that it had connected its first modular cryogenic systems, describing the development as a milestone towards its planned Starling computer in 2029. That remains a roadmap target; the infrastructure work shows the breadth of engineering needed to pursue it. [3] 

My view is that this broader engineering effort deserves more attention in the commercial conversation. A promising qubit becomes valuable through the performance of the system built around it. 

What enterprise buyers should ask 

For organisations assessing quantum technology, this calls for a more demanding conversation with providers. 

What has been demonstrated on hardware? How many reliable operations can the system perform? Does the reported result depend on discarding unsuccessful runs? What resources would be required to tackle a relevant workload, and how would the outcome compare with the strongest available classical approach? 

These questions help distinguish an encouraging experiment from a capability that can support a business decision. 

They also give enterprises a practical way to engage while the hardware develops: identify problems with measurable value, establish credible classical benchmarks and understand which hardware improvements would materially change the calculation. 

A better measure of progress 

At Quantum.Tech, I believe our role is to bring the engineering and enterprise conversations closer together. Researchers need informed customers who can explain the problems worth solving. Enterprise leaders need clear evidence about capabilities, constraints and the work still ahead. 

There is good reason to be encouraged by progress in qubits. Turning that progress into practical computing will require sustained gains in reliability, logical operations and system performance. 

The milestone that matters to a future customer is a useful calculation completed accurately, within a realistic budget and with a clear advantage over the alternatives. That is the standard against which advancing qubits will ultimately be judged. 

Sources 

[1] Google Quantum AI, Quantum error correction below the surface code threshold, Nature. https://www.nature.com/articles/s41586-024-08449-y 

[2] Bluvstein et al., Logical quantum processor based on reconfigurable atom arrays, Nature. https://www.nature.com/articles/s41586-023-06927-3 

[3] IBM, IBM Connects Its First Modular Cryogenic Systems in Milestone Toward Fault-Tolerant Quantum Computing, 19 August 2026. https://newsroom.ibm.com/2026-08-19-ibm-connects-its-first-modular-cryogenic-systems-in-milestone-toward-fault-tolerant-quantum-computing 

Loading