In December 2024, Google announced Willow — a 105-qubit quantum processor that achieved something the company called a "beyond-classical" computation: a benchmark task completed in under five minutes that would take the world's fastest classical supercomputer 10 septillion years (that's 10 followed by 24 zeros). The number is so large it defies intuition, and the announcement triggered immediate debate about what it actually means.
What Makes Willow Different
Previous quantum computers suffered from a fundamental scaling problem: adding more qubits increased error rates faster than it increased computational power, making larger systems less useful rather than more. Willow breaks this pattern. For the first time, Google demonstrated that adding more qubits to their system actually reduces error rates — a key milestone called "below threshold" error correction that the field has been pursuing for twenty years.
The Error Correction Breakthrough
Quantum error correction works by encoding one logical qubit across many physical qubits, using redundancy to detect and correct errors. The catch has always been that the overhead qubits introduce errors of their own, canceling the benefit. Willow's architecture achieves what's called "exponential error suppression" — each time the error-correction code is scaled up, errors drop exponentially rather than increase. This is the threshold that theorists predicted would be necessary for practical quantum computing.
What It Means for Cryptography and Computing
The benchmark Willow solved (random circuit sampling) has no practical application outside of proving quantum advantage. Cryptographically relevant attacks on RSA encryption would require millions of logical qubits — far beyond current capability. But Willow's error correction demonstration is the foundation those systems need. The realistic timeline for cryptographically relevant quantum computing is still measured in decades, but Willow moved the lower bound.
Willow doesn't break encryption. But it proves the theoretical underpinning of quantum error correction works in practice. That's the prerequisite for everything that comes after.
Written by Manas Garge
Founder & Data Engineer