Im Dezember 2024 kündigte Google Willow an — einen neuen Quantenprozessor, der eine bestimmte Benchmark-Berechnung in weniger als fünf Minuten abschloss. Dieselbe Aufgabe würde, so Googles Aussage, die heutigen schnellsten klassischen Supercomputer 10 Septillionen Jahre kosten. Die Ankündigung machte weltweit Schlagzeilen. Aber was bedeutet das eigentlich für Q-Day?
What Willow did — and didn't do
Willow's benchmark was a problem called random circuit sampling. It is specifically designed to be hard for classical computers and easy for quantum ones. It has no known real-world application — it is a test of quantum computational power, not a demonstration of breaking encryption.
Willow cannot run Shor's algorithm against RSA-2048. To do that, researchers estimate a quantum computer would need roughly 4 million physical qubits in a highly error-corrected configuration. Willow has 105 physical qubits. The gap is enormous — but Willow did demonstrate something important: as it added more qubits, its error rate went down rather than up. That had never been achieved before at this scale.
Why error correction changes everything
The central challenge in quantum computing is noise. Qubits are extraordinarily sensitive to their environment — heat, vibration, electromagnetic interference — and errors accumulate rapidly. For years, the working assumption was that adding more qubits would add more errors, making large-scale quantum computers practically impossible to build.
Willow demonstrated below-threshold error correction: a regime where adding more qubits to an error-correcting code actually reduces the overall error rate. This is the theoretical foundation for scaling quantum computers to the sizes needed for cryptographic attacks. It doesn't solve the problem — but it proves the approach works in practice, not just in theory.
What this means for the timeline
Willow does not move Q-Day from 2035 to next year. The engineering challenges that remain — increasing qubit count by four orders of magnitude while maintaining coherence and connectivity — are immense. But it does strengthen the case that the trajectory is real, and that the lower-end estimates (2029–2030) deserve serious attention rather than dismissal.
Google's own security team has publicly stated they are planning for a post-quantum world by 2029. Willow is part of the reason why.