A 30,000-qubit estimate puts exposed bitcoin first in line

Victoria Shinder30 October 2026, 09:402 min read

Quantum estimates for breaking bitcoin's signatures keep falling, and the latest one changes which part of the risk deserves attention. Researchers at the quantum computing company Alice & Bob estimate that a machine with about 29,588 physical qubits could derive a bitcoin private key from its public key in roughly 62 minutes, CoinDesk reported on 29 October (https://www.coindesk.com/tech/2026/10/29/new-quantum-model-slashes-hardware-needed-to-break-bitcoin-keys-by-17-times).

For comparison, Google estimated in March that fewer than 500,000 physical qubits would be needed, down from earlier estimates in the millions. The new figure is about 17 times lower, although the two studies assume different hardware designs. The research is theoretical: no machine of that size exists, and Alice & Bob expects hardware progress and falling resource requirements to meet in the early 2030s.

The timing is the interesting number. At about an hour, the attack is too slow to steal from a transaction while it waits for confirmation, since blocks arrive about every ten minutes; Google's earlier model put that attack at around nine minutes on a much larger machine. Asked whether more qubits could bring the time under ten minutes, senior cryptanalyst Dahmun Goudarzi said the team had not done the calculation, and that there is "no architectural floor".

That leaves the second category of risk: coins whose public keys are already visible on the chain, through address reuse or older output types. More than six million bitcoin sit at such addresses, according to the report. An attacker there has no ten-minute deadline.

What it means

The debate is often framed as a race against confirmation times, but on these numbers the exposed stock is the nearer problem, and it cannot be fixed by faster blocks, only by moving coins to new key types before the hardware arrives. Migration on a decentralised network takes years and needs owners to act, including owners of coins that have not moved in a decade. The estimate comes from a quantum hardware company with an interest in its own architecture; the 17-fold comparison mixes different assumptions and is best read as a direction, not a measurement.