Quantum computers need classical computing to be truly useful

Conventional computers may well be an essential ingredient to making quantum computers truly useful. That’s the message from a gathering of researchers this month, who explained that classical computers are essential for controlling quantum computers, decoding the results of their calculations and even developing new techniques for making quantum computers in the future.

Quantum computers are made from qubits – quantum objects that can take the form of extremely cold atoms or tiny superconducting circuits. The more qubits a quantum computer has, the more computationally powerful it becomes.

But qubits are fragile and therefore must be carefully calibrated, monitored and controlled. Otherwise, they can introduce errors into calculations performed on a quantum computer or render these devices ineffective. To control the qubits, the researchers are turning to classical computing technologies, which they discussed at the AQC25 conference in Boston, Massachusetts, on November 14.

Organized by Quantum Machines, which makes controllers for several types of qubits, the AQC25 conference brought together more than 150 researchers, ranging from quantum computing professors to CEOs of AI start-ups. Across several dozen presentations, they expanded on the role of conventional computing as an enabling technology – and sometimes limiting factor – for the future of quantum computing.

According to Shane Caldwell, a scientist at Nvidia, a fault-tolerant quantum computer for useful problems is only expected to be possible if it is supported by classical computing infrastructure at the petascale – the scale at which the world’s most powerful traditional supercomputers currently operate. Although Nvidia doesn’t make its own quantum computing hardware, the company recently launched a system for connecting quantum computing processors (QPUs) with traditional GPUs – the specialized computing components commonly used in machine learning and high-performance scientific computing.

Even when a quantum computer operates efficiently, its results are presented as a set of quantum properties of its qubits. These must be decoded into more traditional formats to become useful – again requiring conventional computing devices.

Pooya Ronagh of Vancouver-based startup 1Qbit talked about this decoding and how it means the speed of fault-tolerant quantum computers will be determined by the speed at which its classical components, such as controllers and decoders, operate. In other words, whether an expensive machine made from highly specialized quantum hardware needs to run for a few days or a few hours to solve a computational problem may depend on its non-quantum parts.

In another presentation, Benjamin Lienhard of the Walther-Meissner-Institute for Low Temperature Research in Germany explained how using traditional machine learning algorithms could make reading the quantum states of superconducting qubits more efficient. Similarly, Mark Saffman of the University of Wisconsin-Madison reported the use of classical neural networks to improve the readout of qubits made from extremely cold atoms. Regardless of what type of qubit they study, the researchers agreed that non-quantum devices would make these qubits useful.

IBM’s Blake Johnson presented details of the classical computer decoder his team is developing as part of the plan to build a quantum supercomputer by 2029. This supercomputer will use a non-traditional error correction system, and efficient decoding is one of its biggest challenges.

“Over time, we see that the most classic [computing] the closer we get to QPUs, the more we can push the integrated system performance to new limits,” said Yonathan Cohen of Quantum Machines.

Traditional computers even play a role in evaluating how future quantum computers behave and how they will be built. For example, Izhar Medalsy, of a startup called Quantum Elements, said the company’s AI-powered virtual versions of quantum computers — or “digital twins” — can inform actual hardware design.

The Quantum Scaling Alliance, co-led by 2025 Nobel laureate John Martinis, was also represented at the conference. This illustrates the importance of quantum and classical collaboration. The alliance connects qubit builders, traditional IT companies such as Hewlett Packard Enterprise, and materials simulation experts such as software company Synopsys.

The consensus from the conference was clear: the future of quantum computing is fast approaching, but that’s in part thanks to experts who have spent their careers working firmly in the classical world.