20,000 qubits could break Bitcoin’s security, but the required quantum computer does not yet exist

19,397 qubits and 25.7 days

IonQ researchers targeted the 256-bit secp256k1 elliptic curve used by Bitcoin for transaction digital signatures. Its security is based on the difficulty of the elliptic-curve discrete logarithm problem.

Finding a private key from a public key is not realistic with a conventional computer. A sufficiently powerful quantum computer, however, could use Shor’s algorithm to make the same task solvable.

According to the new work, this would require a customised trapped-ion quantum computer with 19,397 physical qubits. One calculation would take an estimated 25.7 days. Based on the researchers’ estimate, the probability of a single attempt succeeding reaches around 63 percent.

That final figure matters. “26 days” does not mean a guaranteed private key by the evening of day 26, but the estimated duration of one quantum computation.

The major breakthrough lies in the qubit count

What makes the work notable above all is the scale of the required hardware. Previous estimates have required substantially more physical qubits for tasks in the same class.

The IonQ team simultaneously optimised the algorithm, compiler, error correction and a quantum-computing architecture called Walking Cat. The final calculation uses around 1,450 logical qubits and approximately 40 million Toffoli gates.

The researchers did not stop at an abstract algorithm. They compiled the calculation down to the level of a specific architecture, taking into account qubit movement, error correction, ion loss and their replacement, among other factors.

As a result, 20,000 qubits is a far more meaningful estimate than simply the theoretical claim that Shor’s algorithm may one day break Bitcoin cryptography.

Bitcoin cannot currently be attacked this way

This is where the sensational headline diverges from reality. IonQ itself states explicitly that no current quantum computer can carry out the described attack. The study did not break any wallet, private key or part of the Bitcoin network.

The problem is not just the qubit count. It requires nearly 20,000 sufficiently high-quality physical qubits with working error correction, as well as a system capable of performing an enormous number of quantum operations continuously for almost a month.

The study even assumes continuous ion replacement, as some are lost from the trap during operation. The calculation indicates that the system would need to add an average of around 10 ions per second throughout the entire 26-day run. That is not quite the same as leaving a laptop to break Bitcoin overnight.

The threat concerns private keys, not “cracking open” the Bitcoin blockchain

Another important clarification: a quantum computer would not erase the Bitcoin blockchain or magically calculate all coins for itself.

Shor’s algorithm would attack the cryptography behind digital signatures. If an attacker obtains a user’s public key and can derive the private key from it quickly enough, they could in principle create a valid signature and move the corresponding funds.

This means the issue is not limited to Bitcoin. Elliptic-curve cryptography is also widely used beyond cryptocurrency. That is precisely why the study matters more than another story about Bitcoin’s possible demise. The question concerns an entire generation of modern public-key cryptography.

26 days could shrink quickly in the future

The study’s 25.7 days is not a limit imposed by any law of physics. It is the result of a specific architecture and specific assumptions.

The authors have already reduced the time required for the most expensive operation in a single calculation by roughly 31 times compared with the more general solution in the Walking Cat architecture. Without those optimisations, the same calculation would have taken an estimated 411 days.

From a cybersecurity perspective, this is the study’s most uncomfortable aspect. The threat does not depend only on how quickly the qubit count of quantum computers grows. Improvements in error correction, algorithms and architecture are equally important. Better mathematics could make existing hardware far more dangerous without increasing the qubit count tenfold.

There is no reason to panic now, but there is reason to migrate

The study does not show that Bitcoin will become useless in 26 days, or even 26 months. Instead, it shows the class of machine that cryptography must be able to withstand in the future.

IonQ itself links the calculation to the need to transition to post-quantum cryptography. Quantum-resistant algorithms already exist, and the challenge is adopting them early enough.

For Bitcoin, this means that in the longer term its signing solution will need to be changed before a cryptographically relevant quantum computer is actually built.