New estimates From Google Quantum AI show quantum attack on Bitcoin is closer than thought

Google Quantum AI researchers have determined that breaking the core cryptography of various cryptocurrencies, secured by the secp256k1 curve, may require as few as 1200 logical qubits and 90 million Toffoli gates, a significantly lower threshold than previously understood. The team’s work elucidates specific vulnerabilities blockchain technologies face with the development of quantum computers and potential mitigation strategies. To ensure responsible disclosure, the researchers validated their findings using a zero-knowledge proof without revealing specific attack vectors. This analysis reveals that emerging “fast-clock” quantum computers could enable attacks on cryptocurrency transactions in the public mempool.

Shor’s Algorithm Estimates for secp256k1 Bitcoin Attacks

This represents a significant reduction in the estimated resources needed for a successful attack compared to earlier projections, bringing the threat of quantum decryption closer to reality. These architectures, the researchers note, could enable “on-spend” attacks targeting public mempool transactions, potentially allowing malicious actors to seize funds before they are confirmed on the blockchain.

A key distinction highlighted in the analysis is the difference between fast-clock and “slow-clock” quantum computers, such as those based on neutral atoms or ion traps. The researchers found that circuits executing Shor’s algorithm on superconducting architectures, with a 10^-3 physical error rate and planar connectivity, could complete the calculation in minutes using fewer than half a million physical qubits.