Existing crypto would be broken, but quantum computation also seems provide much stronger encryption possibilities by leveraging entanglement states. [1]

1: https://en.m.wikipedia.org/wiki/Quantum_cryptography

Not all existing crypto would be broken. Symmetric encryption algorithms like AES will be fine. Asymmetric encryption (public-key) algorithms like RSA/DSA will be in trouble due to Shor's algorithm, but other public-key systems will be fine. Hash functions like SHA-2 and SHA-3 will be mostly fine, because Grover's algorithm is at best sub-exponential, not quadratic.

Also, quantum cryptography mostly provides superior key establishment/management capabilities, not superior confidentiality guarantees. It's a misnomer to state that we can achieve "stronger encryption" by "leveraging entanglement states"; most of quantum computing is useful for cryptanalytic attacks right now, not cryptographic construction.

Thanks, that's very enlightening!

The next version of ethereum includes a "bring your own cryptography algo" feature that lets you substitute your preferred algorithm for the default choice. This gives it safeguards against the outside risk that a direct attack against core cryptography ever becomes feasible.

Merkle–Damgård algorithms are thought to be safe. I'm not sure about Keccak.

Keccak (SHA-3) is currently thought to be safe. The current state of the art would be Brassard-Hoyer-Tapp[1], but SHA3-512 should be secure against that. It's basically a birthday attack that generates a giant table, then uses Grover's algorithm to find a collision.

_______

1. https://link.springer.com/chapter/10.1007%2FBFb0054319