EIP-8355: Precompiles for ML-DSA Verification: When Post-Quantum Cryptography Meets Ethereum's Gas Economics

The proposal appeared on the Ethereum Magicians forum in September 2026 with the sober authority of a NIST standard being imported into a blockchain ecosystem that had spent the previous decade optimizing for speed rather than longevity. EIP-8355 wou

· Updated September 16, 2026 · Zain Tran · 11 min read · 2 total views · 2 today

Categories: technology

Featured image for EIP-8355: Precompiles for ML-DSA Verification: Whe

The proposal appeared on the Ethereum Magicians forum in September 2026 with the sober authority of a NIST standard being imported into a blockchain ecosystem that had spent the previous decade optimizing for speed rather than longevity. EIP-8355 would add three precompiled contracts to the Ethereum Virtual Machine for verifying ML-DSA signatures—the post-quantum cryptographic standard that NIST had finalized in 2024 and that government agencies were already beginning to require for sensitive communications. The pitch was future-proofing: quantum computers capable of breaking ECDSA and RSA were still theoretical, but the migration to quantum-resistant cryptography would take years, and Ethereum needed to start now. The fine print raised a question about whether a blockchain that struggles to process current transactions efficiently should be adding cryptographic primitives that make every signature verification ten times more expensive.

That was the proposal. Then came the question of whether post-quantum cryptography on a chain that already costs too much to use is foresight or just an expensive way to solve a problem that does not exist yet while making the problems that do exist worse.

What EIP-8355 Actually Proposes

The standard describes three precompiled verification contracts with specific parameters:

The ML-DSA Precompiles:

- ML-DSA-44: NIST security level 1, equivalent to AES-128, smallest signature and key sizes

- ML-DSA-65: NIST security level 3, equivalent to AES-192, medium signature and key sizes

- ML-DSA-87: NIST security level 5, equivalent to AES-256, largest signature and key sizes

The Verification Operations:

- Signature verification: Confirm that a message was signed by the claimed public key

- Public key validation: Verify that a public key is well-formed and meets security parameters

- Batch verification: Verify multiple signatures in a single operation for gas efficiency

The NIST Context:

- FIPS 204: The NIST standard that specifies ML-DSA (Module Lattice-based Digital Signature Algorithm)

- PQC migration: Federal mandate for post-quantum cryptography adoption by 2035

- CRYSTALS-Dilithium: The lattice-based scheme that ML-DSA is based on

- Security levels: Standardized strength categories based on computational hardness assumptions

The Ethereum Integration:

- Precompile addresses: Reserved EVM addresses for the three verification functions

- Gas costs: Priced to reflect computational complexity (~10-50x more expensive than ECDSA)

- Solidity support: New opcode wrappers for contract developers

- Backward compatibility: Existing ECDSA signatures continue to work; ML-DSA is additive

The proposal frames these as infrastructure investments. They are also a bet that Ethereum will still exist and matter when quantum computers become a threat.

alt text

Key Metrics at a Glance

Parameter ECDSA (Current) ML-DSA-44 ML-DSA-65 ML-DSA-87 Impact
Public Key Size 33 bytes 1,312 bytes 1,952 bytes 2,592 bytes Much larger
Signature Size 65 bytes 2,420 bytes 3,293 bytes 4,595 bytes Much larger
Verification Gas ~3,000 ~30,000 ~45,000 ~65,000 10-20x higher
Security Level Classical NIST Level 1 NIST Level 3 NIST Level 5 Quantum-resistant
Standardization SECP256K1 FIPS 204 FIPS 204 FIPS 204 Government-backed
Adoption Timeline Live since 2015 Draft/precompiles Draft/precompiles Draft/precompiles Years away
Quantum Threat Vulnerable Resistant Resistant Resistant Future-proof

The Proprietary Quantum Readiness Efficiency Score (QRES)

I've developed a framework to evaluate whether adding ML-DSA precompiles is a wise investment for Ethereum:

Formula: QRES = (Threat Imminence × 0.25) + (Migration Readiness × 0.25) + (Gas Impact Sustainability × 0.2) + (Ecosystem Demand × 0.15) + (Implementation Quality × 0.15)

EIP-8355 Assessment:

Factor Score Analysis
Threat Imminence 2/10 Quantum computers capable of breaking ECDSA do not exist yet; optimistic estimates suggest 10-20 years before cryptographically relevant quantum computers; pessimistic estimates suggest 5-10 years; the threat is real but not imminent; Ethereum has time to prepare but not infinite time
Migration Readiness 4/10 Adding precompiles is the easy part; migrating the entire Ethereum ecosystem from ECDSA to ML-DSA is the hard part; every wallet, every exchange, every smart contract that verifies signatures would need updates; the migration would take years and cost billions in engineering effort
Gas Impact Sustainability 3/10 ML-DSA verification costs 10-20x more gas than ECDSA; on a chain where gas costs are already prohibitive for many users, this additional cost would make signature verification prohibitively expensive for casual transactions; the gas impact is sustainable only if L2 scaling dramatically reduces costs first
Ecosystem Demand 2/10 As of September 2026, there is no production demand for ML-DSA on Ethereum; government and institutional users who require post-quantum cryptography are not using Ethereum for those applications; the demand exists in theory but not in practice
Implementation Quality 7/10 The precompiles are based on well-reviewed NIST standards; the CRYSTALS-Dilithium algorithm has received extensive cryptanalysis; the implementation risk is low if the precompile code is properly audited; the technical quality is sound
Total QRES 3.6/10 The proposal is technically sound but premature; the threat is distant, the migration burden is enormous, the gas impact is severe, and the demand does not exist yet

A score of 3.6 indicates that EIP-8355 is a high-quality solution to a problem that Ethereum does not yet need to solve at the cost of making its current problems worse.

alt text

The Three Post-Quantum Precompile Traps

Trap 1: The Gas Cost Explosion

Ethereum's gas economics are already strained. A simple token transfer costs ~21,000 gas. A complex DeFi interaction can cost 200,000-500,000 gas. Adding ML-DSA verification at 30,000-65,000 gas per signature does not just increase costs—it changes the economics of every application that verifies signatures. A multi-sig wallet that requires three signatures goes from ~9,000 gas to ~90,000 gas. A rollup that batches thousands of signatures sees its verification overhead increase by an order of magnitude. The chain that was supposed to become more accessible becomes more expensive for the applications that actually exist today, in order to protect against a threat that exists only in theory.

Trap 2: The Migration Mirage

The argument for adding ML-DSA precompiles now is that migration takes time and Ethereum should start early. But starting early does not mean finishing early. The Ethereum ecosystem has a poor track record of coordinated upgrades. The transition to proof-of-stake took seven years. The adoption of ERC-4337 account abstraction is still minimal. The shift to Layer 2 is ongoing and incomplete. Adding ML-DSA precompiles does not create a migration path—it just creates a new option that developers can ignore while continuing to use ECDSA. The migration that is supposed to happen over a decade may not happen at all because the incentives to migrate do not exist until the threat is imminent, and by then it may be too late.

Trap 3: The Opportunity Cost Neglect

Every engineering hour spent on ML-DSA precompiles is an engineering hour not spent on problems that Ethereum faces today. The chain still has MEV extraction that disadvantages ordinary users. The rollup ecosystem still has fragmented liquidity and poor interoperability. The gas costs still price out non-wealthy users. The validator set still has concentration risks. The wallet experience is still confusing for newcomers. Adding post-quantum cryptography to address a threat that is 10-20 years away while these immediate problems remain unsolved is an allocation of resources that the ecosystem may regret. The opportunity cost of foresight is the neglect of the present.

Competitive Landscape: Post-Quantum Cryptography Adoption

Blockchain/Platform PQ Strategy Implementation Cost Impact Migration Status QRES
Ethereum (proposed) Precompiles ML-DSA (FIPS 204) 10-20x gas Not started 3.6/10
Bitcoin None planned None None N/A N/A
Cardano Research phase N/A N/A Research 4.0/10
Algorand Falcon signatures Live ~5x Limited 5.5/10
Quantum-resistant L1s Native PQ Various Baseline Complete 8.0/10
Signal (messaging) CRYSTALS-Kyber Live Minimal Complete 8.5/10
Cloud providers NIST standards Hybrid modes Transparent In progress 7.0/10

The comparison shows that successful post-quantum migration happens either in systems where PQ is native (new chains) or in centralized systems where the migration can be mandated (Signal, cloud providers). Ethereum's decentralized, opt-in model makes migration uniquely difficult.

alt text

Scenario Analysis: Three Futures for Ethereum Post-Quantum Cryptography

Scenario A: Coordinated Migration (20% probability)

  • Quantum threat timeline accelerates, forcing ecosystem-wide urgency
  • Major wallets and exchanges commit to ML-DSA migration timelines
  • Layer 2 scaling reduces gas costs sufficiently to absorb ML-DSA overhead
  • Ethereum becomes a leader in decentralized post-quantum infrastructure
  • QRES improves to 6.5/10

Scenario B: Stalled Adoption (55% probability)

  • ML-DSA precompiles are added but ecosystem adoption remains minimal
  • ECDSA continues to dominate because migration costs exceed perceived threat
  • Quantum computers remain theoretical longer than anticipated
  • The precompiles become unused infrastructure, maintained but irrelevant
  • QRES degrades to 2.5/10

Scenario C: Emergency Scramble (25% probability)

  • Quantum computing breakthrough happens suddenly
  • Ethereum ecosystem panics and attempts rapid migration
  • The migration is rushed, poorly coordinated, and creates security vulnerabilities
  • Users who do not migrate lose funds to quantum-enabled attackers
  • QRES becomes irrelevant as survival takes priority

The Bottom Line

EIP-8355 is technically excellent work from cryptographers who understand the quantum threat. The Quantum Readiness Efficiency Score is 3.6/10. Threat imminence is low. Migration readiness is poor. Gas impact is unsustainable. Ecosystem demand does not exist. Implementation quality is high.

The three traps—gas cost explosion, migration mirage, and opportunity cost neglect—are structural risks that accompany every attempt to future-proof infrastructure against distant threats. They reflect the fundamental tension between long-term security and short-term usability. The community that wants to survive the quantum era must also accept that survival preparations today may kill the patient before the disease arrives.

The deeper question is whether Ethereum should be a general-purpose platform that solves every problem or a focused platform that solves the problems its users actually have. Post-quantum cryptography is important for the world. It may not be important for Ethereum in 2026. The federal contractor who needs FIPS 204 compliance is not using Ethereum for signature verification. The DeFi trader who pays $50 in gas for a swap is not asking for larger signatures that cost more to verify. The NFT collector who just wants to prove ownership is not worried about quantum computers.

The proposal deserves recognition for technical correctness. The researchers who drafted EIP-8355 are not imagining the threat; they are responding to a real one that has been documented by NIST and acknowledged by cryptographers worldwide. But technical correctness is not the same as priority. The Ethereum ecosystem that adds ML-DSA precompiles today is an ecosystem that has chosen to optimize for a distant future while its present continues to deteriorate. The precompile that verifies quantum-resistant signatures at 65,000 gas is a monument to foresight that the users paying those gas costs cannot afford.

TL;DR

  • What: EIP-8355 proposes three EVM precompiled contracts for verifying ML-DSA (Module Lattice-based Digital Signature Algorithm) signatures at NIST security levels 1, 3, and 5, importing the FIPS 204 post-quantum standard into Ethereum
  • The Score: Quantum Readiness Efficiency Score of 3.6/10—threat imminence (2/10) is distant; migration readiness (4/10) is poor for ecosystem-wide coordination; gas impact sustainability (3/10) is severe with 10-20x higher verification costs; ecosystem demand (2/10) does not exist yet; implementation quality (7/10) is technically sound
  • The Reality: Post-quantum cryptography is important for the world but not urgent for Ethereum in 2026; the gas cost explosion makes current transactions more expensive to protect against a theoretical future threat
  • Three Traps: Gas cost explosion (10-20x signature verification overhead strains already expensive gas economics); migration mirage (adding precompiles does not create migration path—ecosystem has poor track record of coordinated upgrades); opportunity cost neglect (engineering resources spent on distant threat instead of immediate problems like MEV, fragmentation, and wallet UX)
  • Outlook: Coordinated migration (20%) with accelerated quantum threat and ecosystem urgency; stalled adoption (55%) where precompiles exist but ECDSA dominance continues; emergency scramble (25%) where sudden quantum breakthrough forces rushed, vulnerable migration

Sources


Zain Tran is TotesTek's Ethereum Ecosystem Columnist & Accountability Reporter. He writes about Ethereum, ETH, smart contracts, DeFi, Layer 2 networks, staking, validators, and the real-world consequences of technical and financial failure.