Quantum Computer Surpasses the '100-Qubit Barrier'... Challenges Google and IBM [Reading Science]

Quantinuum Unveils 98-Qubit Trapped Ion Processor

A New Phase in Quantum Competition: Scaling Up While Maintaining Accuracy

The quantum computing industry has reached another significant milestone. The trapped ion method, which has so far been recognized for its high accuracy but struggled to scale, has unveiled a system close to 100 qubits for the first time, issuing a serious challenge to the superconducting quantum computer camp led by Google and IBM.

On June 18, the international journal Nature published the research results on "Helios," a 98-qubit trapped ion quantum processor developed by the U.S. quantum computing company Quantinuum.

Reference photo to aid understanding of the article. In June last year, at the Korea Standard Science Institute booth during 'Quantum Korea 2025' held at the aT Center in Seocho-gu, Seoul, a visitor is examining a model of a superconducting 50-qubit quantum computer. Photo by Yonhap News Agency

Helios is a trapped ion quantum computer that utilizes barium ions (Ba+) confined within an electromagnetic field as qubits. The research team applied a rotatable ion storage ring and a Quantum Charge-Coupled Device (QCCD) structure to achieve "all-to-all" connectivity, enabling every qubit to be directly connected to every other qubit.

The most significant aspect of this study is that it greatly increased scale while maintaining accuracy.

The quantum computer development race has two main pillars: expanding the number of qubits and reducing errors. The superconducting approach has excelled at rapidly increasing qubit numbers but suffers from relatively high error rates, while the trapped ion approach maintains high accuracy but has found large-scale expansion difficult.

Kim Kihwan, Head of the Trapped Ion Quantum Science Group at the Institute for Basic Science (IBS), commented, "The trapped ion camp has aimed to increase scale while preserving maximum accuracy. This research demonstrates that the trapped ion method, which had remained relatively small in scale, can now implement nearly 100 qubits."

Will It Overcome the "Quantum Advantage" Debate?

The research team reported that their benchmark experiments with Helios achieved an average two-qubit gate fidelity of 99.921%.

In particular, they claimed to have provided more convincing evidence of quantum advantage through Random Circuit Sampling (RCS) experiments.

Quantum advantage refers to the state in which a quantum computer outperforms existing supercomputers in solving specific problems.

Reference photo to help understand the article. Fujitsu quantum computer. The Asia Business Daily DB

In 2019, Google declared the world's first quantum advantage using a superconducting quantum computer, but some researchers raised controversy by suggesting that classical computers could potentially simulate the process using alternative calculations.

Director Kim noted, "Google's early experiments had a success probability of about 0.1%, but this study achieved approximately 5%. This demonstrates much more convincingly that quantum systems can surpass classical computers on specific problems."

Professor Han Jeonghun of the Department of Physics at Sungkyunkwan University commented, "Until now, systems with around 100 qubits were considered the domain of superconducting technology. This study shows that ion-based systems can implement quantum circuits of equivalent scale."

"Accurate but Slow"... Commercialization Remains a Challenge

Experts point out that while this achievement is a meaningful step forward, there are still many obstacles to commercialization.

The trapped ion method requires moving and cooling qubits, resulting in relatively slow computation speeds. While the superconducting method operates on the microsecond (μs) scale, the trapped ion method operates on the millisecond (ms) scale.

Kwon Seokjun, Professor of Semiconductor Convergence Engineering at Sungkyunkwan University, explained, "This study demonstrates the expansion potential of the trapped ion method, but in order to achieve large-scale error correction and commercialization, issues such as computation speed and optical interconnection between modules still need to be addressed."

Reference photo to aid understanding of the article. Quantum computer model. Provided by Yonhap News Agency

Nevertheless, there are significant implications for Korea as well.

Experts advise that, rather than jumping directly into the quantum hardware race like the United States and China, Korea should focus on supply chain competitiveness in areas where it has strengths—such as semiconductor manufacturing, control chips, and photonic integration technology.

Professor Kwon said, "Trapped ion chips are manufactured based on semiconductor processes. Korea can secure sufficient competitiveness in semiconductor manufacturing, control electronics, and quantum–artificial intelligence (AI) hybrid infrastructure."

This content was produced with the assistance of AI translation services.