Cryptosecurity about to break?

Background

This post discusses the research paper Securing Elliptic Curve Cryptocurrencies against Quantum Vulnerabilities: Resources Estimates and Mitigations, which recently appeared in the leading physics journal PRX Quantum [1].

Published by Google Quantum AI, the Ethereum Foundation, and Stanford University, this theoretical paper estimates that future quantum computers can break standard blockchain cryptocurrency encryption in minutes using about twenty times fewer physical hardware resources than was previously thought.

Conclusion: Current cryptocurrency stashes might soon be unsafe. This also has implications for government, industrial security, privacy, etc.

While the resources required are still quite a bit far away from coming together, I thought the paper was worth discussing since scientific/technological advances sometimes come in spurts (disruptively) and not incrementally. In other words, what is orders of magnitude away from possibility today might become a daily reality in a short while. So we need to be prepared.

Disclaimer: I am not invested in any cryptocurrencies, nor have any association with entities who run them.

What the Technical Discussion Implies

Private keys used by cryptocurrencies like Bitcoin and Ethereum rely on a technique called 'the Elliptic Curve Discrete Logarithm Problem (ECDLP-256) over the secp256k1 curve' for their security. To break this encryption using a classical computer would take billions of years. Basically, it is not practicable, and so our secrets are safe from such an attack.

However, it can be breached by a sufficiently powerful quantum computer using Shor’s algorithm in a much shorter time. The question is what resources are required to make this happen. The authors found that Shor’s algorithm can run using about 1200 logical qubits and 80 million Toffoli gates. In practical terms, this translates to about 10 minutes of computational time on 500,000 physical qubits. This is about 20 times lower than previously estimated.

What do we have available at present? About 1000 physical qubits and 100 Toffoli gates. Sounds like there's no need to worry. But experts project a suitable quantum computer which could have access to such resources in a few years. Moreover, there are parallel estimates which bring the numbers down to 26,000 physical qubits, with a computational time of a few days [2]. Conclusion: the threshold is getting closer all the time.

One detail I found interesting was the distinction made by the authors on the basis of hardware speeds. They identify architectures with fast clocks (like superconducting or photonic quantum computers) which are quick enough to complete an attack in several minutes (and redirect funds during a transaction, for example; this has been given the name of `on-spend attack'). On the other hand, there are architectures with slow clocks (trapped-ion and neutral atom quantum computers), which are too sluggish to disrupt network traffic in real time.

Specific Blockchain Vulnerabilities

The study looks at different network designs and their specific risks:

• Bitcoin: Uses secp256k1 signatures. Any address that has exposed its public key (such as through a previous outgoing transaction) is at risk. Old coins left in early wallets are also major targets. 

• Ethereum: Uses the BLS12-381 curve, which is quantum-vulnerable. If an attacker steals keys from more than 1/3 of the network validators (specialized nodes or computers in a blockchain network responsible for checking transactions), they can freeze the network. If they control more than 2/3, they can rewrite transaction history. 

  
  

Responsible Disclosure

Quantum researchers face a difficult choice. Sharing full attack circuit details helps science, but it also hands malevolent parties a weapon. Releasing nothing makes it difficult to establish credibility. 

To solve this problem, the authors used the `SP1 Zero-Knowledge Virtual Machine'. They created a cryptographic proof showing their attack circuits work within the stated resource limits without publishing the dangerous proprietary blueprints. This proof was verified using a Groth16 zk-SNARK, a method for proving something without disclosing the underlying data. 

Conclusions

The authors strongly urge all blockchain communities to transition to Post-Quantum Cryptography (PQC) right away. PQC uses new mathematical formulations that quantum computers cannot easily break. Institutions such as NIST have been releasing PQC standards for some time now, mandatory for federal systems in the US, and also adopted elsewhere in the world.

Beyond code upgrades, the paper suggests new public policies called 'digital salvage'. These rules would help communities safely manage or destroy long-abandoned dormant assets, preventing hackers from stealing coins left behind by inactive users. 

Stay tuned for further updates!

[1] PRX Quantum 7, 031001 (2026).

[2] arXiv:2603.28627.