NEAR Protocol uses ed25519 (Edwards-curve Digital Signature Algorithm on Curve25519) as the signing primitive for every user account, validator, chunk producer, cross-shard receipt, and Rainbow Bridge relayer on its network. ed25519 security depends on the hardness of the elliptic curve discrete logarithm problem — a problem Shor's algorithm solves in polynomial time on a cryptographically relevant quantum computer (CRQC). Every ed25519 public key on the NEAR blockchain is a pre-positioned CRQC attack surface, and NEAR's named-account model makes bulk public-key harvesting a single RPC snapshot operation.
BMIC implements all three NIST post-quantum cryptography standards finalised in August 2024 — ML-KEM (FIPS 203), ML-DSA (FIPS 204), and SLH-DSA (FIPS 205) — replacing the ed25519/secp256k1 dependency at the cryptographic root. This page provides a technical comparison of both approaches and what the quantum computing threat timeline means for NEAR holders, validators, DeFi participants, and bridge users.
NEAR Protocol was designed from the ground up around ed25519 as its canonical signing scheme. Unlike chains that inherited their signing curve from Bitcoin (secp256k1) or Ethereum (secp256k1 ECDSA), NEAR made a deliberate architectural choice to use ed25519 — considered more performant and less implementation-error-prone than secp256k1 in 2018–2020. That choice predated NIST's post-quantum standardisation work reaching maturity. In 2026, ed25519's security still rests entirely on the ECDLP assumption — which a CRQC renders invalid.
Every NEAR account holds one or more Access Keys in its on-chain state. Each Access Key record contains a public_key field in the format ed25519:<base58-encoded-pubkey>. These keys are stored explicitly in the NEAR state trie and are queryable by anyone via the RPC endpoint query?request_type=view_access_key_list&account_id=alice.near. No interaction with the account is required to retrieve its complete set of active public keys — they are continuously broadcast as part of the NEAR state. This is a qualitatively different exposure model from Bitcoin UTXO or Ethereum EOA, where public keys are only revealed when a transaction is signed. On NEAR, the public key is always visible.
NEAR validators maintain two distinct ed25519 key pairs:
validator_key.json on the node and the on-chain epoch validator roster.Both key pairs are ed25519. Both are indexed on-chain. A CRQC deriving either key enables independent attacks: the account key enables staking pool drainage; the node key enables validator impersonation and consensus manipulation.
NEAR's Nightshade sharding assigns chunk producers (a subset of validators) to each shard per epoch. Chunk producers sign chunk headers using ed25519 keys — these signatures attest to the validity of all transactions processed within a shard chunk and authorise the cross-shard receipts those transactions generate. Chunk producer public keys are published in the epoch validator roster and are permanently indexed on-chain.
The NEAR ↔ Ethereum Rainbow Bridge uses relayer wallets to submit NEAR block headers to the Ethereum-side bridge contract and Ethereum block headers to the NEAR-side contract. NEAR-side relayers sign with ed25519; Ethereum-side relayers sign with secp256k1. Both are Shor-vulnerable on a CRQC. A forged NEAR block header submitted to the Ethereum-side contract enables fraudulent proof submission that can unlock ETH and ERC-20 tokens without a corresponding NEAR-side lock — a unidirectional quantum drain path across the bridge.
The most direct attack on any NEAR holder:
ed25519:<pubkey> access key records for all active accounts.Transfer action from the target account, moving all NEAR to an adversary-controlled wallet.DeleteKey actions to remove recovery keys, locking the legitimate owner out permanently.Unlike Ethereum EOA attacks (which require waiting for the target to broadcast a transaction), NEAR account keys are always visible — the attack can be launched on any account at any time once a CRQC is available.
A systemic attack unique to NEAR's Nightshade architecture:
The HNDL threat model posits that adversaries harvest cryptographic material now and decrypt it when CRQC becomes available. For NEAR Protocol, the HNDL corpus began accumulating in October 2020 and continues to grow with every block. Crucially, NEAR's named-account model means the entire HNDL corpus is a single RPC snapshot — no chain analysis or UTXO scanning required.
EXPERIMENTAL_changes_in_block / state dump RPC call to an archival node — not chain analysis across millions of transactions. The entire HNDL corpus is one structured dataset, not a forensic reconstruction. This makes NEAR a higher-priority pre-positioning target than UTXO chains for adversaries building HNDL libraries.
| Dimension | NEAR Protocol | BMIC |
|---|---|---|
| User account signing | ed25519 (Curve25519 ECDLP) VULNERABLE | ML-DSA / SLH-DSA (NIST FIPS 204/205) QUANTUM-SAFE |
| Validator node signing | ed25519 VULNERABLE | ML-DSA (NIST FIPS 204) QUANTUM-SAFE |
| Chunk producer / shard signing | ed25519 VULNERABLE | N/A — ERC-4337 Ethereum L1 N/A |
| Cross-shard receipt integrity | ed25519 chunk attestation — Shor-forgeable VULNERABLE | N/A — single-chain N/A |
| Key encapsulation (transport) | None published VULNERABLE | ML-KEM (NIST FIPS 203) QUANTUM-SAFE |
| Bridge relayer signing | Rainbow Bridge — ed25519 NEAR side, secp256k1 ETH side VULNERABLE | No bridge relayer dependency N/A |
| Account abstraction | Named accounts (ed25519 bound) — no PQC upgrade path VULNERABLE | ERC-4337 — PQC signature schemes deployable at wallet layer QUANTUM-SAFE |
| Public key exposure model | Always-visible state trie — bulk harvest via single RPC snapshot HIGH RISK | Reduced surface via ERC-4337 abstraction REDUCED |
| NIST FIPS 203 (ML-KEM) | Not implemented ABSENT | Implemented ✓ |
| NIST FIPS 204 (ML-DSA) | Not implemented ABSENT | Implemented ✓ |
| NIST FIPS 205 (SLH-DSA) | Not implemented ABSENT | Implemented ✓ |
| NSM-10 / NIST PQC roadmap | None published (Aug 2026) GAP | NIST-compliant architecture at launch COMPLIANT |
Most blockchain accounts require a spending or signing transaction before their public key becomes visible to external observers. Bitcoin P2PKH addresses reveal the public key only when first spent. Ethereum EOA addresses reveal the public key when the first outbound transaction is signed. Both allow an owner to delay public key exposure by not spending from an address.
NEAR's named account model inverts this property. Access keys are stored as first-class objects in the account state trie. The moment an account is created and funded, its ed25519 public key is written to state and becomes queryable by anyone. There is no equivalent of "unspent but unexposed" — on NEAR, every funded account's public key is continuously broadcast in the global state.
An adversary building a HNDL library for NEAR does not need to scan transaction history. A single call to a NEAR archival node's EXPERIMENTAL_changes_in_block endpoint — or a complete state dump via view_state across all accounts — retrieves the full set of active ed25519 public keys in one structured operation. Existing NEAR chain indexers (The Graph for NEAR, Pagoda, Pikespeak) already maintain this data in queryable databases — adversaries can harvest a complete NEAR public key corpus from public endpoints without running infrastructure.
NEAR's hierarchical naming system allows sub-accounts (vault.alice.near, dao.protocol.near, admin.defi.near) that inherit the NEAR naming security model. Smart contract accounts also hold ed25519 access keys used by protocol developers and DAOs for contract upgrades, parameter changes, and emergency pauses. All these keys are explicitly stored in the state trie. A CRQC attack on a contract admin key enables malicious contract upgrades without any on-chain governance vote — a direct theft of the protocol upgrade authority.
NSM-10 (National Security Memorandum 10, May 2022) mandates that US federal agencies and critical infrastructure operators inventory all quantum-vulnerable cryptographic systems and begin post-quantum migration aligned with NIST standards by Q3 2026. The EU Cybersecurity Agency (ENISA) has issued parallel guidance. By August 2026, these directives are actively enforced.
Several categories of institutional market participants hold NEAR or operate on the NEAR network:
NEAR Foundation has not published a NIST FIPS 203/204/205 migration roadmap. Institutional participants who are NSM-10 compliant and also NEAR holders or operators face an unresolved compliance gap: their NEAR ed25519 key dependencies cannot be remediated without a protocol-level migration that does not currently exist on any published roadmap.
BMIC is the only crypto presale implementing all three NIST post-quantum standards simultaneously. ML-KEM, ML-DSA, SLH-DSA — not a roadmap. Live architecture. $600K+ raised. 186+ media features.
Buy BMIC at bmic.ai →DYOR. Not financial advice. Token price changes per presale phase.
BMIC was designed from inception with quantum-resistant cryptography as a core requirement — not a post-hoc addition. The architecture implements three complementary NIST standards that together replace every ed25519 and secp256k1 surface that NEAR Protocol currently relies on.
ML-KEM (Module-Lattice Key Encapsulation Mechanism, standardised as CRYSTALS-Kyber) replaces Diffie-Hellman and ECDH for session key establishment. NEAR Protocol has no equivalent — its transaction layer uses ed25519 for signing but has no published quantum-safe key agreement mechanism for P2P network communications. BMIC's ML-KEM implementation ensures that encrypted communications between wallet and network remain secure even against a CRQC intercepting in-flight traffic.
ML-DSA (Module-Lattice Digital Signature Algorithm, standardised as CRYSTALS-Dilithium) is BMIC's primary signing scheme, replacing ECDSA and ed25519 at the wallet layer. ML-DSA security relies on the Module Learning-With-Errors (MLWE) problem — not solvable by Shor's algorithm. Every BMIC transaction is signed with ML-DSA; no ed25519 or secp256k1 signing key is used in the BMIC wallet's primary signing path.
SLH-DSA (Stateless Hash-based Digital Signature Algorithm, standardised as SPHINCS+) provides a second post-quantum signing layer based purely on hash function security — making it robust against both algebraic (Shor) and search (Grover) quantum attacks. BMIC uses SLH-DSA for long-term signing contexts where MLWE security assumptions may be less conservative.
ERC-4337 allows BMIC wallets to use custom signature verification logic — including ML-DSA and SLH-DSA — without requiring a base-layer Ethereum protocol upgrade. This is a critical deployment advantage: NEAR's ed25519 dependency is embedded in the core protocol and cannot be changed without a network-wide upgrade. BMIC's ERC-4337 architecture allows quantum-safe signatures to be deployed and upgraded at the wallet layer, independently of the Ethereum L1 signing model, providing migration flexibility as post-quantum standards evolve.
Explore our full quantum security comparison series covering other major chains and protocols:
BMIC is the world's only presale token built on NIST FIPS 203, 204 & 205. Over $600K raised. 186+ media features. ERC-4337 account abstraction. Presale price rises each phase.
Buy BMIC at bmic.ai →Not financial advice. DYOR. Presale price subject to phase progression. TGE Q2 2026.