Google's quantum computing division has reported that its Willow processor achieved a computational milestone that classical supercomputers would require approximately 10,000 years to replicate — completing the task in under four minutes. The result, published in Nature, represents a significant advance beyond Google's 2019 quantum supremacy demonstration and, more importantly, demonstrates a property that researchers have been chasing for two decades: error correction that improves as the system scales.

The announcement landed with the weight of genuine scientific significance. Unlike the 2019 result — which was immediately contested by IBM, who argued their classical supercomputers could solve the same problem in days — the Willow result has been independently scrutinised and the margin of supremacy is orders of magnitude larger. The scientific community's response has been notably less sceptical this time.

What Was Actually Solved

The specific problem — a variant of random circuit sampling — is not directly useful in itself. It involves sampling from the output distribution of a random quantum circuit, a task that is computationally hard for classical computers but natural for quantum ones. The problem was chosen precisely because it is well-understood theoretically, making it possible to verify the quantum result and to calculate how long a classical computer would take to replicate it.

The 10,000-year figure refers to the best known classical algorithm running on the world's most powerful supercomputer. It is not a claim that no classical algorithm could ever solve the problem faster — quantum supremacy claims are always relative to the best known classical methods, and classical algorithms do improve over time. But the margin is large enough that closing it classically would require a breakthrough of comparable magnitude to the quantum advance itself.

More significantly, Willow demonstrates a key property that has eluded quantum computers: error correction that improves as the system scales. Previous quantum processors became less reliable as more qubits were added — errors accumulated faster than they could be corrected. Willow shows the opposite trend. As Google added more qubits to the error-correction code, the logical error rate decreased exponentially. This is the behaviour that quantum error correction theory predicts, but that no physical system had previously demonstrated convincingly.

Why Error Correction Is the Real Breakthrough

To understand why this matters, it helps to understand the fundamental challenge of quantum computing. Quantum bits (qubits) are extraordinarily sensitive to environmental disturbances — heat, electromagnetic interference, even cosmic rays can cause errors. Classical computers deal with errors through redundancy and error-correcting codes, but applying the same approach to quantum computers is much harder because quantum states cannot be copied (the no-cloning theorem) and measuring a quantum state destroys it.

Quantum error correction works by encoding a single logical qubit across many physical qubits, in a way that allows errors to be detected and corrected without measuring the logical state directly. The theory has been understood since the 1990s, but implementing it in practice requires physical qubits with error rates below a threshold — and maintaining that threshold as the system scales has proven extremely difficult.

Willow's demonstration that error rates decrease as the system scales is the first convincing evidence that a physical quantum system can operate below the error correction threshold at scale. It does not mean fault-tolerant quantum computing is imminent — the error rates are still too high for most practical applications — but it demonstrates that the path to fault tolerance is open. The question is no longer whether scalable quantum error correction is physically possible, but how long it will take to reach the error rates required for useful computation.

Implications for Cryptography

The cryptography community has been watching quantum computing progress closely, because sufficiently powerful quantum computers could break RSA and elliptic curve encryption — the foundations of most internet security. Willow's result has reignited that conversation, though the immediate threat is less acute than some headlines suggest.

"We're not there yet," says Dr. Michele Mosca of the Institute for Quantum Computing. "But today's result is a clear signal that the timeline for cryptographically relevant quantum computing needs to be taken seriously. Organisations that haven't started migrating to post-quantum cryptography should start now."

Breaking RSA-2048 — the most common encryption standard — would require a fault-tolerant quantum computer with millions of logical qubits. Willow has 105 physical qubits. The gap is enormous. But the demonstration of scalable error correction means the gap is now a matter of engineering rather than fundamental physics. The NIST post-quantum cryptography standards, finalised in 2024, provide the migration path that organisations need to follow — and the Willow result is a compelling argument for urgency.

The most vulnerable systems are those with long data lifetimes. Encrypted data intercepted today could be stored and decrypted later, once quantum computers are powerful enough. Intelligence agencies and adversarial nation-states are almost certainly doing exactly this — a strategy known as "harvest now, decrypt later." For data that needs to remain confidential for 10 or more years, the migration to post-quantum cryptography is not optional.

Drug Discovery and Molecular Simulation

The more immediately promising application is molecular simulation. Quantum computers are naturally suited to modelling quantum mechanical systems — including the behaviour of molecules. Accurate molecular simulation could dramatically accelerate drug discovery by allowing researchers to test potential drug candidates computationally before ever synthesising a single molecule.

Current classical computers can accurately simulate molecules up to about 50 atoms. Beyond that, the computational complexity grows exponentially and approximations become necessary. Many of the most important biological molecules — proteins, enzymes, drug targets — are far larger than this limit. A fault-tolerant quantum computer could simulate these molecules exactly, potentially identifying drug candidates that classical methods would miss entirely.

The timeline for this application is longer than the cryptography threat. Molecular simulation requires fault-tolerant quantum computers with thousands of logical qubits — still years away. But pharmaceutical companies including Roche, Pfizer, and Merck are already investing in quantum computing research, positioning themselves to exploit the technology when it matures.

Materials Science and Climate Applications

Beyond drug discovery, quantum simulation could transform materials science. Designing better catalysts for industrial chemical processes — including the Haber-Bosch process that produces fertiliser, which accounts for roughly 2% of global energy consumption — could have enormous economic and environmental impact. Designing better solar cell materials, better battery chemistries, and better superconductors are all problems that quantum simulation could accelerate.

The climate implications are potentially significant. Many of the most important clean energy technologies — next-generation solar cells, room-temperature superconductors, better electrolysers for green hydrogen production — are limited by our inability to design materials with precisely the right quantum mechanical properties. Quantum computers could remove that limitation.

The Competitive Landscape

Google is not alone in the race to build useful quantum computers. IBM has its own roadmap, targeting 100,000 physical qubits by 2033. Microsoft is pursuing a fundamentally different approach using topological qubits, which are theoretically more resistant to errors. IonQ and Quantinuum are building trapped-ion quantum computers, which have lower error rates than superconducting qubits but are harder to scale. China's quantum computing programme, centred at the University of Science and Technology of China, has demonstrated its own supremacy results.

The diversity of approaches reflects genuine uncertainty about which physical platform will ultimately prove most practical for fault-tolerant quantum computing. Each approach has different trade-offs between qubit quality, scalability, and operating conditions. The field is still early enough that the winning platform is not obvious.

What Comes Next

Google's roadmap calls for a fault-tolerant quantum computer — one capable of running practical algorithms with error rates low enough for real applications — by the end of the decade. Whether that timeline holds will depend on continued progress in error correction, qubit quality, and the classical control systems that manage quantum processors.

The Willow result is a genuine milestone, not a marketing announcement. It demonstrates that the fundamental physics of scalable quantum error correction works as theory predicts. The engineering challenges that remain are formidable, but they are engineering challenges — the kind that yield to sustained investment and effort. The era of practically useful quantum computing is closer than it was last week.

Sources & Further Reading