In a milestone that pushes quantum computing toward practical use, Google has announced the first demonstration of verifiable quantum advantage—a long-sought benchmark showing a quantum computer solving a problem beyond the reach of classical machines, with results that can be independently checked.
At the heart of this advance is Willow, Google’s superconducting quantum chip, which ran an algorithm called Quantum Echoes 13,000 times faster than the world’s top supercomputers. The achievement, published in Nature, represents a rare moment when both hardware and software progress aligned to deliver measurable, reproducible results.
This verifiability, that Quantum Echoes can be repeated on another quantum platform, marks an essential bridge from laboratory experiments to reliable computation.
Willow: A New Benchmark in Quantum Hardware
Built on superconducting circuits that function as artificial atoms, Willow operates across a 105-qubit array with extraordinary precision: 99.97% fidelity for single-qubit operations and 99.88% for entangling gates. That level of accuracy enabled Google’s researchers to perform millions of measurements within seconds, of more than one trillion in total over the course of the project so far.
The chip’s performance stems from decades of work in quantum circuit design, culminating in breakthroughs by John Clarke, John Martinis, and Michel Devoret, who were awarded the 2025 Nobel Prize in Physics for their foundational research. Devoret, now Google’s chief scientist of quantum hardware, helped lead the current experiment.
The Quantum Echoes Algorithm
The Quantum Echoes algorithm works by sending a signal into a quantum system, then reversing the process to detect an echo that reveals the system’s hidden dynamics, essentially allowing researchers to listen to quantum behavior. Using Willow, scientists ran sequences of entangled qubits forward and backward, observing interference patterns that classical machines cannot simulate.
To ensure accuracy, Google’s team red-teamed its results, probing and challenging them, applying years’ worth of equivalent classical simulations to test robustness. These independent verification steps confirmed that the algorithm’s outputs cannot be matched by any existing classical supercomputer.
Early Signs of Real-World Use
While the experiment itself was not designed for commercial use, the same approach has immediate real-world scientific implications. In a companion study, Google scientists applied the algorithm to simulate molecules used in nuclear magnetic resonance spectroscopy, which is the study of how matter interacts with electromagnetic radiation. The test uncovered new atomic details in two molecules, producing findings that could one day inform advances in drug design and materials science.
Yet to reach this ambitious goal, quantum hardware will need to grow roughly 10,000-fold in scale, Google estimates. Achieving such a leap requires both hardware improvements and smarter error-correction algorithms, a focus of Google’s Quantum AI Roadmap. The company has already reached what it calls the third milestone on that roadmap, demonstrating below-threshold error correction, and is now targeting its next: the creation of a long-lived logical qubit, a building block for large-scale systems.
Within Five Years?
Industry experts hailed Google’s advance while noting that optimism should be tempered: building a fully scalable, fault-tolerant quantum system remains on the horizon, perhaps the distant horizon.
Still, the pace of quantum’s advance has clearly picked up in the last year or so. Google’s result not only showcases technical mastery but strengthens the case for superconducting qubits as the architecture most likely to scale. The company certainly has rivals in quantum: IBM, Microsoft, and a number of smaller players (Rigetti, D-Wave, IonQ) are pursuing their own approaches, yet Google’s latest work sets an impressive benchmark.
If predictions hold, useful quantum applications, in areas like drug discovery, energy storage, and material design, could emerge within five years. For now, Willow’s performance suggests that the dream of reliable, verifiable quantum computing has shifted from theory to tangible progress.
