LPsLux Proposals
LP-536

Seal Finality via Quantum Event Horizon

Draft

Post-quantum finality guarantees for data integrity seals through dual-certificate consensus

Category
Core
Created
2026-02-13

Abstract

This LP defines the finality semantics for data integrity seals (LP-0535) anchored through the Z-Chain Receipt Registry (LP-0530). A seal achieves quantum-final status when the block containing its receipt passes the Quasar consensus Event Horizon — the point at which dual classical (BLS) and post-quantum (lattice) certificates have been aggregated. Once quantum-final, a seal cannot be reversed by any adversary, including one with access to a cryptographically relevant quantum computer.

This LP specifies the finality states, the transition rules, and the on-chain interface for querying seal finality status.

Motivation

The Finality Gap

Data integrity seals are only as trustworthy as the finality of the chain they're anchored to. Existing approaches anchor to:

  • Bitcoin: Probabilistic finality. A 6-block confirmation (~60 minutes) provides high confidence but never mathematical certainty. A sufficiently motivated state actor with >50% hash power can reorg.
  • Ethereum: Committee-based finality (~13 minutes to finalization). Relies on BLS signatures over BN254 — vulnerable to quantum attack via Shor's algorithm on the pairing group.

Neither provides the guarantee that high-stakes use cases demand:

Use CaseRequirementBTC/ETH Finality
Court-admissible evidenceMathematically irreversibleInsufficient
Regulatory compliance (EU AI Act)Provably tamper-proofTime-dependent
Whistleblower protectionCannot be reversed by state actorsInsufficient
Insurance claimsSurvives dispute periods (years)Quantum-vulnerable
Medical recordsLong-term integrity (decades)Quantum-vulnerable

The Quantum Threat to Seal Integrity

When a cryptographically relevant quantum computer (CRQC) exists:

  1. ECDSA breaks → Bitcoin transaction signatures can be forged → anchored roots can be replaced
  2. BLS breaks → Ethereum validator attestations can be faked → finalized blocks can be reversed
  3. Every seal anchored to BTC/ETH loses its guarantee

This is not a theoretical concern. NIST has standardized post-quantum algorithms (FIPS 203-205) specifically because the timeline for CRQC is measured in years, not decades. Seals created today must remain valid for decades.

Lux Solution: Quantum Event Horizon

The Quasar consensus engine provides a finality mechanism that survives quantum attacks: the Event Horizon. A seal that passes the Event Horizon has dual-certificate finality — both classical and post-quantum signatures attest to its inclusion in the canonical chain. No future adversary, regardless of computational capability, can reverse it.

Specification

Seal Finality States

A data integrity seal transitions through four finality states:

┌──────────┐    ┌───────────┐    ┌──────────────┐    ┌────────────────┐
│ PENDING  │───>│ CLASSICAL │───>│   QUANTUM    │───>│ HORIZON-FINAL  │
│          │    │  FINAL    │    │    FINAL     │    │                │
│ In mempool│    │ BLS cert  │    │ BLS + Lattice│    │ Checkpointed   │
│ or block │    │ only      │    │ dual cert    │    │ in Event       │
│          │    │           │    │              │    │ Horizon        │
└──────────┘    └───────────┘    └──────────────┘    └────────────────┘
     │                │                │                     │
   ~0s             ~500ms          ~3s                    ~6s

State 1: PENDING

The seal transaction has been submitted but not yet included in a block, or is in a block that has not received sufficient validator votes.

Security: No finality guarantee. Transaction can be dropped or reordered.

State 2: CLASSICAL_FINAL

The block containing the seal's receipt has received a BLS aggregate signature from ≥2/3 of validators by stake weight.

Security: Byzantine fault tolerant under classical assumptions. Safe against any adversary without a quantum computer. Equivalent to Ethereum's finality.

Time: ~500ms from block proposal.

State 3: QUANTUM_FINAL

The block (or its containing quantum bundle) has received both:

  • BLS aggregate signature (classical certificate)
  • Lattice-based aggregate signature via Ringtail protocol (post-quantum certificate)

Security: Safe against any adversary, including one with a CRQC. The lattice certificate uses Ring-LWE based signatures requiring ≥2^160 operations to forge.

Time: ~3 seconds from block proposal (quantum bundles aggregate every ~3 seconds).

State 4: HORIZON_FINAL

The quantum-final block has been included in an Event Horizon checkpoint. The checkpoint itself has been signed by ≥2/3 of validators with both classical and post-quantum signatures, and the checkpoint references all prior checkpoints in a hash chain.

Security: Maximum finality. Even if the entire validator set is replaced, the checkpoint chain provides an independent proof of finality. Equivalent to a "point of no return" — no future consensus execution can produce a conflicting history.

Time: ~6 seconds from block proposal (checkpoints every ~6 seconds).

On-Chain Interface

// SPDX-License-Identifier: MIT
pragma solidity ^0.8.25;

/// @notice Seal finality states
enum SealFinality {
    Pending,          // Not yet finalized
    ClassicalFinal,   // BLS certificate only
    QuantumFinal,     // BLS + Lattice dual certificate
    HorizonFinal      // Checkpointed in Event Horizon
}

/// @notice Query seal finality status
interface ISealFinality {
    /// @notice Get the finality state of a seal
    /// @param receiptId The receipt ID from LP-0530
    /// @return finality Current finality state
    /// @return classicalAt Block number when classical finality was achieved (0 if pending)
    /// @return quantumAt Block number when quantum finality was achieved (0 if not yet)
    /// @return horizonAt Block number when horizon finality was achieved (0 if not yet)
    function sealFinality(uint256 receiptId) external view returns (
        SealFinality finality,
        uint256 classicalAt,
        uint256 quantumAt,
        uint256 horizonAt
    );

    /// @notice Check if a seal has achieved at least the specified finality level
    /// @param receiptId The receipt ID
    /// @param required Minimum required finality level
    /// @return met Whether the requirement is met
    function meetsFinalityRequirement(
        uint256 receiptId,
        SealFinality required
    ) external view returns (bool met);

    /// @notice Get the latest Event Horizon checkpoint
    /// @return height Checkpoint height
    /// @return root Merkle root of all receipts up to this checkpoint
    /// @return classicalSig BLS aggregate signature
    /// @return quantumSig Lattice aggregate signature
    function latestHorizon() external view returns (
        uint256 height,
        bytes32 root,
        bytes memory classicalSig,
        bytes memory quantumSig
    );

    /// @notice Get the Event Horizon checkpoint containing a specific seal
    /// @param receiptId The receipt ID
    /// @return checkpointHeight The checkpoint that includes this seal
    /// @return inclusionProof Merkle proof from receipt to checkpoint root
    function horizonProof(uint256 receiptId) external view returns (
        uint256 checkpointHeight,
        bytes32[] memory inclusionProof
    );
}

Precompile address: 0x0536 (adjacent to Data Seal Registry at 0x0535)

Finality Requirements by Use Case

Applications SHOULD specify their minimum finality requirement:

Use CaseMinimum FinalityRationale
Real-time monitoringPendingSpeed over finality
Supply chain trackingClassicalFinalAdequate for near-term
Legal evidenceQuantumFinalMust survive decades
Court submissionHorizonFinalMaximum provability
Insurance claimsQuantumFinalMulti-year dispute periods
Regulatory complianceHorizonFinalAudit trail must be absolute
Whistleblower evidenceHorizonFinalState-level adversary
AI model attestationQuantumFinalRegulatory longevity

Dual-Certificate Structure

Each quantum bundle contains:

QuantumBundle {
    height:        uint64              // Bundle height
    blockRange:    [startBlock, endBlock]  // Classical blocks covered
    merkleRoot:    bytes32             // Poseidon2 root of all receipts in range
    classicalCert: BLSAggregate {
        signature:   [96]byte          // BLS12-381 aggregate
        signerBits:  bitfield          // Which validators signed
        weight:      uint64            // Total stake weight
    }
    quantumCert:   LatticeCert {
        signature:   []byte            // ~3KB Ringtail aggregate
        signerBits:  bitfield          // Which validators signed
        weight:      uint64            // Total stake weight
        epoch:       uint64            // Key epoch (for rotation tracking)
    }
}

A bundle is valid when:

  • classicalCert.weight >= 2/3 * totalStake
  • quantumCert.weight >= 2/3 * totalStake
  • merkleRoot correctly commits to all receipts in blockRange

Event Horizon Checkpoint Chain

Checkpoints form an append-only chain:

Checkpoint {
    height:        uint64
    parentHash:    bytes32             // Hash of previous checkpoint
    bundleRange:   [startBundle, endBundle]
    accumulatorRoot: bytes32           // Poseidon2 root of ALL receipts ever
    classicalCert: BLSAggregate
    quantumCert:   LatticeCert
    timestamp:     uint64
}

checkpointHash = Poseidon2(
    DST_CHECKPOINT,
    height,
    parentHash,
    accumulatorRoot,
    timestamp
)

The checkpoint chain is independently verifiable: given any checkpoint and the accumulator state, a verifier can confirm the entire history without replaying consensus.

Cross-Chain Finality Export

For seals that must be verified on external chains (Ethereum, other L1s), the horizon proof can be wrapped in a Groth16 proof:

HorizonExportProof {
    receiptHash:       bytes32    // The specific receipt
    inclusionProof:    bytes32[]  // Merkle path to accumulator root
    checkpointHeight:  uint256    // Which checkpoint
    accumulatorRoot:   bytes32    // Root at checkpoint time
    groth16Proof:      bytes      // ~256 bytes, verifiable on any EVM
}

The Groth16 proof attests: "This receipt is included in a Poseidon2 Merkle accumulator whose root was signed by ≥2/3 of Lux validators with both classical and post-quantum certificates."

This proof is verifiable on any EVM chain via a single staticcall to a Groth16 verifier (~200K gas).

Rationale

Why Dual Certificates Instead of Post-Quantum Only?

Lattice-based signatures are larger (~3KB vs 96 bytes for BLS) and slower to verify. Using lattice signatures alone would:

  1. Increase block size significantly
  2. Slow down consensus
  3. Lose the performance benefit of BLS aggregation

The dual approach provides:

  • Fast classical finality (~500ms) for latency-sensitive applications
  • Quantum finality (~3s) for long-term applications
  • Graceful degradation — if lattice schemes are found vulnerable, classical certificates still hold (and vice versa)
  • Migration path — when quantum computers arrive, the system continues operating on lattice certificates alone

Why Checkpoint Chain?

The Event Horizon checkpoint chain provides finality proofs that are:

  1. Self-contained — no need to replay consensus from genesis
  2. Compact — a single checkpoint + Merkle proof suffices
  3. Independently verifiable — any third party can verify without running a node
  4. Long-lived — checkpoints remain valid forever (no expiration)

Why Poseidon2 for Checkpoint Hashing?

Poseidon2 is used throughout the checkpoint chain (not just the receipt accumulator) because:

  1. ZK-friendly — checkpoint proofs can be wrapped in ZK proofs for cross-chain export
  2. Post-quantum safe — the checkpoint chain itself doesn't depend on DLP assumptions
  3. Consistent — same hash function at every layer reduces implementation complexity and attack surface

Backwards Compatibility

This LP extends LP-0530 (Receipt Registry) and LP-0535 (Data Integrity Seal) without breaking changes. Existing receipts and seals automatically receive finality tracking — the finality state is derived from consensus, not from the receipt itself.

Applications that do not query finality status are unaffected.

Test Cases

Finality State Transitions

  1. Seal created → state = Pending
  2. Block receives BLS aggregate (≥2/3 stake) → state = ClassicalFinal
  3. Quantum bundle includes block, receives lattice aggregate → state = QuantumFinal
  4. Checkpoint includes bundle → state = HorizonFinal

Edge Cases

  1. Block reorg before classical finality — seal returns to Pending, re-included in new block
  2. Lattice key rotation mid-bundle — bundle uses keys from the epoch that started it
  3. Validator set change between classical and quantum finality — quantum certificate uses the validator set at bundle creation time
  4. Cross-chain export of horizon proof — Groth16 proof verifiable on Ethereum mainnet

Performance Targets

MetricTarget
Time to ClassicalFinal<500ms
Time to QuantumFinal<3s
Time to HorizonFinal<6s
Checkpoint proof size<2KB
Groth16 export proof size<256 bytes
Export proof verification gas (EVM)<200K gas

Security Considerations

Quantum Adversary Model

The primary adversary is a state-level actor with access to a CRQC capable of breaking:

  • ECDSA (secp256k1, secp256r1)
  • BLS12-381
  • Any pairing-based cryptography

Against this adversary:

  • ClassicalFinal seals are NOT safe — BLS certificates can be forged
  • QuantumFinal seals are safe — lattice certificates require ≥2^160 operations
  • HorizonFinal seals are maximally safe — checkpoint chain provides independent verification

Signature Agility

The dual-certificate design provides cryptographic agility. If either scheme is found vulnerable:

  • BLS broken (quantum attack): Lattice certificates alone provide finality. Classical finality state is skipped.
  • Lattice broken (cryptanalytic advance): BLS certificates provide finality. Lattice scheme is replaced via governance.
  • Both broken: Consensus halts safely (no unsafe finality granted). New signature scheme deployed via hard fork.

Long-Term Seal Validity

Seals at HorizonFinal remain valid indefinitely because:

  1. The Poseidon2 accumulator root is self-proving (no external dependency)
  2. The dual certificates are independently verifiable
  3. The checkpoint chain forms an immutable append-only log
  4. Cross-chain Groth16 exports create standalone proofs

Even if the Lux Network itself ceased to operate, any party holding a horizon proof can independently verify the seal's integrity.

References

Implementation

Quasar Consensus Engine

Repository: github.com/luxfi/consensus

DirectoryComponent
protocol/quasar/Core Quasar engine with quantum bundling
protocol/quasar/quantum_block.goAsync quantum block production
protocol/quasar/epoch.goLattice key rotation and epoch management
protocol/horizon/Event Horizon checkpoint algorithms
core/dag/horizon.goDAG-level horizon interface

FHE Seal Examples

Repository: github.com/luxfi/fhe

ExampleDescription
cmd/sealBoolean-circuit document sealing with XNOR+AND verification
cmd/provenanceAI model provenance via encrypted hash comparison
cmd/mediasealMedia content authentication and tamper detection

Copyright and related rights waived via CC0.