Context
Public Ethereum Foundation research in 2026 is moving Ethereum L1 away from Poseidon toward mature hashes such as SHA / BLAKE, enabled by binary-field proving designs (“hash-friendly SNARKs”). That signal matters. It is not identical to Zeqron’s proving stack.
What Zeqron uses today
| Surface | Primitive | Role | |---------|-----------|------| | Protocol IDs / contracts | ParallelHash (SP 800-185) | Domain-separated protocol hashing | | DID / QronId | BLAKE3 | Identity display and related surfaces | | STARK / Hologram / PQ-VRF | Poseidon2 (BabyBear, width=32 per RFC-STARK-001) | Circuit-native commitment hash | | Storage MAC | KMAC256 | Authenticated storage | | QronHandshake KDF | HKDF-SHA3-256 | Session key derivation |
Poseidon2 is not a universal protocol hash on Zeqron.
Decision
Retain Poseidon2 for scoped STARK paths while monitoring hash-friendly STARK migrations and EF binary-field progress. Substitution is tracked via crypto-agility (RFC-CRYPTO-PQ-005). Replacing Poseidon2 requires a proving-path migration, not a one-line swap.
Non-claims
- ::We do not claim Ethereum is “wrong.”
- ::We do not claim Poseidon2 has SHA/BLAKE-class public cryptanalysis maturity.
- ::Transparent STARKs do not, by themselves, prove end-to-end post-quantum security of the whole L1.
Full decision text and maturity grids: Security Profile.
