Scientific breakthrough gives new hope to building quantum computers

Experts in quantum computing believe a significant technical challenge in creating practical quantum computers has been overcome, potentially paving the way for operational full-scale systems by the end of this decade.

This optimism follows Google’s announcement of achieving a major breakthrough in addressing quantum systems' inherent instability. The findings, which initially circulated informally in August, were formally published in the peer-reviewed journal Nature on Monday.

Google also revealed details of its advanced quantum chip, which was used to achieve this milestone. The company claims this innovation will help scale its technology toward practical applications.

Researchers likened Google’s achievement to the first man-made nuclear chain reaction in 1942—a theoretical prediction that became a reality after years of technological advancements. William Oliver, a physics professor at MIT, remarked that this quantum breakthrough had been theoretically envisioned in the 1990s, and the results have been long anticipated. Scott Aaronson, a computer science professor at the University of Texas at Austin, echoed this sentiment, noting that engineering often takes decades to catch up with theoretical ideas.

A persistent hurdle in quantum computing has been creating stable systems capable of large-scale computations. Quantum computers rely on phenomena like superposition, where particles can exist in multiple states simultaneously, and entanglement, where quantum particles influence each other. However, qubits—the fundamental units of quantum computing—can only maintain their quantum states for fractions of a second, causing information to quickly decay.

As computations involve more qubits and operations, errors accumulate due to increased "noise." Scientists have sought to address this using error correction techniques, which encode information across multiple qubits to preserve coherence even as individual qubits fail. However, effective error correction requires high-quality qubits that collectively produce meaningful results.

In their Nature paper, Google researchers claimed to have crossed this critical threshold. By scaling from a 3x3 grid of qubits to 5x5 and 7x7, they reported a consistent halving of errors at each step. Julian Kelly, Google’s director of quantum hardware, described the findings as robust evidence that their system could overcome inherent instability, a vital step toward scaling up to thousands of qubits for useful computations.

Hartmut Neven, head of quantum at Google, explained that the next goals include further reducing error rates and linking multiple qubit clusters for practical computations. He attributed these advancements to steady hardware improvements, including a switch to manufacturing qubits in-house. Google’s new qubits can maintain their quantum states for nearly 100 microseconds—five times longer than its earlier hardware.

Improved qubit stability, combined with advanced production methods, is expected to lower costs. Google aims to reduce component costs tenfold by 2030, estimating the price of a full-scale quantum system at around $1 billion.

Despite these advancements, competitors have raised concerns about Google’s approach. IBM, which is also vying to develop the first fault-tolerant quantum computer, questioned the scalability of Google’s surface code for error correction. Jay Gambetta, IBM’s quantum computing head, argued that surface code might require billions of qubits for practical computations, prompting IBM to adopt a modular design requiring fewer qubits. However, IBM’s approach, involving a three-dimensional qubit arrangement, introduces its own challenges, such as developing new connectors, which it plans to address by 2026.

Google remains confident in its strategy. Neven stated that the company’s research demonstrated scalability, estimating that a full-scale quantum system would require approximately 1 million qubits.