Detail
The threat model
Public-key cryptography relies on math problems (factoring, discrete log) that classical computers cannot solve efficiently. Quantum computers can solve them, given enough qubits — the timeline is 5-15 years depending on the estimate. Meanwhile, harvest-now-decrypt-later attacks are already collecting encrypted data to break once quantum arrives.
The standards
NIST ran a Post-Quantum Cryptography competition from 2016 to 2024. The winners: ML-KEM (Kyber) for key encapsulation, ML-DSA (Dilithium) for digital signatures, SLH-DSA (SPHINCS+) for stateless hash-based signatures. Falcon is standardized as FN-DSA for tight signature sizes.
How H33 uses it
Every H33 signature is ML-DSA-65 (Dilithium at 128-bit security level). Every key exchange is ML-KEM. Biometric matching happens under Fully Homomorphic Encryption — the template never decrypts, not even to H33. Benchmark v12 measures the full lifecycle at 2.29 million authentications per second on commodity hardware.
Why this matters now
The transition from classical to post-quantum crypto is a multi-year architectural change. Systems that started PQ-native (like H33) get five years head start over systems that need to retrofit under deadline. When quantum lands, retrofit systems will be doing that while under attack.