# Post-Quantum Zero-Knowledge Proofs Overview

**Source**: https://quantumsequrity.com/blog/post-quantum-zero-knowledge
**Category**: Future Research

---

[← Back to Blog](../../blog) Future Research

# Post-Quantum Zero-Knowledge Proofs Overview

11 min read

A zero-knowledge proof is one of the most counterintuitive ideas in modern cryptography. The basic claim is that you can prove a statement is true without revealing why. The classic example: you can prove you know the password to your account without telling the system what the password is. The system gets convinced; the password stays secret.

For decades, zero-knowledge proofs were theoretical. In the last ten years they have become practical, and they now power blockchain rollups, privacy-preserving authentication systems, and verifiable computation. Most of these implementations rely on cryptographic assumptions that a future quantum computer would break. The good news is that several flavors of zero-knowledge proof are already post-quantum secure. This piece walks through what zero-knowledge proofs are, where post-quantum versions stand, and what to expect over the next decade.

## A Plain-English Introduction

Imagine you have a password and you want to log in to a website. The traditional approach: you send the password to the server, the server hashes it and compares to its stored hash. The password leaves your computer in some form, and the server learns enough to recompute the comparison.

Zero-knowledge authentication works differently. You and the server perform a back-and-forth where you demonstrate you know the password without ever sending it. The server gets convinced. The password never leaves your computer.

Generalize this to any statement. "I am the holder of an account at this bank." "This transaction is valid under these rules." "I am over 18." Each can be proved without revealing the underlying data. The proof itself is short, and it can be checked by anyone with the right verification key.

This is the basic capability. The implementations vary in size, speed, and underlying cryptographic assumptions.

## SNARKs and Their Cryptographic Assumptions

SNARK stands for Succinct Non-Interactive Argument of Knowledge. The "succinct" part means the proof is small, often just a few hundred bytes regardless of the statement being proved. The "non-interactive" part means the proof can be generated and verified without a back-and-forth conversation. SNARKs are the most widely deployed kind of zero-knowledge proof today, especially in blockchain applications.

Traditional SNARKs (like Groth16 or PLONK with KZG commitments) rely on pairing-based cryptography. Pairings are mathematical operations on elliptic curves that have a particular structure. A future quantum computer running Shor's algorithm could break the underlying problems, rendering these SNARKs insecure.

Several research efforts are building post-quantum SNARKs. The most promising approaches use lattice-based commitments instead of pairing-based ones, or replace pairings with hash-based commitments that are quantum-resistant. As of 2026, post-quantum SNARKs are not yet at the same performance level as their pairing-based counterparts, but the gap is narrowing.

For more on the underlying threat to classical primitives, see [why-rsa-2048-will-break](why-rsa-2048-will-break.md).

## STARKs: Already Post-Quantum

STARK stands for Scalable Transparent Argument of Knowledge. STARKs were proposed by Eli Ben-Sasson, Iddo Bentov, Yinon Horesh, and Michael Riabzev in 2018. The proof system relies only on hash functions and arithmetic over finite fields. Both of these are post-quantum secure.

STARKs differ from SNARKs in two important ways. First, STARK proofs are larger. A typical STARK proof is tens to hundreds of kilobytes versus a few hundred bytes for a SNARK. Second, STARKs do not require a trusted setup. SNARKs based on pairings often need a setup ceremony where some random values are generated and then destroyed; if the random values leak, the entire system can be compromised. STARKs avoid this by being "transparent": all randomness is publicly derivable.

STARKs are deployed in production today. StarkWare's StarkEx and StarkNet are layer-2 scaling solutions on Ethereum that use STARKs to compress thousands of transactions into a single onchain proof. Polygon's zkEVM and other systems have similar architectures.

For applications that already use STARKs, post-quantum migration is essentially complete on the proof system side. The application's other cryptographic primitives may still need migration (for example, signatures on transactions), but the zero-knowledge layer is already PQ-safe.

## Lattice-Based Zero-Knowledge

Lattice-based cryptography underlies the NIST PQC standards (ML-KEM and ML-DSA). The same lattice problems can be used to build zero-knowledge proofs, with the same post-quantum security properties.

Lattice-based ZK schemes have made significant progress in the last few years. Schemes like Bulletproofs+ on lattices, lattice-based PLONK variants, and dedicated constructions like Banquet and Picnic show that lattices can produce competitive proofs.

The tradeoff is proof size. Lattice-based proofs tend to be larger than pairing-based SNARKs but smaller than STARKs. They sit in a middle ground: typically a few kilobytes per proof.

For more on lattice-based primitives in general, see [ml-kem-explained](ml-kem-explained.md).

## Hash-Based Zero-Knowledge

Hash-based cryptography is the most conservative branch of post-quantum cryptography. It relies only on the security of cryptographic hash functions. SLH-DSA, the NIST hash-based signature standard, is in this family.

Hash-based zero-knowledge schemes use the same primitives. They tend to produce the largest proofs of any post-quantum approach but the strongest security guarantees. If a hash-based ZK scheme is compromised, it means the underlying hash function (like SHA-256 or SHA-3) was broken, which would be a much larger cryptographic event than just compromising the ZK system.

For applications that prioritize conservative security over compact proofs, hash-based ZK is attractive. For most consumer-facing use cases, the proof size penalty is too high.

For more on hash-based signatures, see [lms-stateful-signatures](lms-stateful-signatures.md).

## Use Cases for Post-Quantum Zero-Knowledge

Zero-knowledge proofs have applications well beyond blockchain. Some that benefit specifically from post-quantum versions:

Long-lived identity attestations. Suppose a government issues a digital credential that proves citizenship. The credential lasts for years. If the underlying ZK proof is built on classical cryptography, a future quantum computer could forge attestations retroactively. A post-quantum ZK system avoids this risk.

Verifiable voting. Cryptographic voting protocols use ZK to prove that votes were counted correctly without revealing individual votes. The integrity of historical elections matters indefinitely. PQ ZK ensures that future quantum capability does not retroactively undermine election results.

Compliance proofs for finance. A financial institution might need to prove that a portfolio meets regulatory requirements without revealing the portfolio's contents. The proof might need to remain verifiable for years (audit trails for tax purposes). PQ ZK provides this durability.

Verifiable AI inference. As AI models become more powerful, the ability to prove that an output came from a specific model running on specific inputs becomes valuable. ZK schemes for verifying ML inference are an active research area, and PQ versions are part of the broader landscape.

For broader PQC application context, see [what-is-post-quantum-cryptography](what-is-post-quantum-cryptography.md).

## Trusted Setups and Their PQ Story

A trusted setup is a one-time procedure that generates parameters used by a ZK scheme. In pairing-based SNARKs (like Groth16), the setup produces a structured reference string that everyone uses. The setup involves random values that must be destroyed afterward; if they leak, the system is broken.

Setup ceremonies are a cryptographic and operational achievement. The Zcash powers-of-tau ceremony involved hundreds of independent participants, each contributing entropy and then provably destroying their secret. As long as at least one participant was honest, the setup is secure.

In a post-quantum world, trusted setups remain meaningful but their cryptographic guarantees may weaken. The structured reference string is itself a cryptographic artifact, and its security might depend on assumptions a quantum computer could violate. Migrating away from pairing-based SNARKs eliminates this concern.

STARKs and many lattice-based schemes do not require a trusted setup. This is one of their attractive properties.

## Performance and Practicality

The current state of post-quantum zero-knowledge can be summarized roughly:

STARKs are production-ready and post-quantum secure. They have larger proof sizes than classical SNARKs but acceptable verifier performance.

Lattice-based SNARKs are advancing rapidly in research. Practical deployment is starting in specialized applications.

Hash-based ZK is conservative and works but is rarely the right choice for performance-sensitive applications.

The biggest performance gap relative to classical cryptography is in prover time. Generating a STARK or lattice-based proof can take seconds or minutes for moderate computations, compared to fractions of a second for some pairing-based SNARKs. This gap is narrowing as research advances and hardware improves.

## Recursion and Aggregation

A powerful idea in modern ZK is recursion: a proof can verify another proof. This lets you compress many small proofs into a single one, or chain proofs across many computations.

Recursion has been demonstrated for both SNARKs and STARKs. Post-quantum recursive proof systems are an active research area. As they mature, they enable use cases like compressing an entire blockchain history into a single succinct proof, or enabling verifiable computation that can run for days but is verified in milliseconds.

The performance characteristics of recursive PQ proofs are still being figured out. The first generation of practical recursive PQ systems is expected within a few years.

## Standardization Status

Unlike traditional cryptography, zero-knowledge has limited formal standardization. NIST's post-quantum effort focused on key encapsulation and signatures, not zero-knowledge. There is no equivalent of FIPS 203 for ZK.

The closest standardization efforts are:

The IETF's IRTF Crypto Forum Research Group occasionally publishes informational RFCs about ZK schemes. These are documentation rather than mandatory standards.

ZKProof, an industry consortium, publishes recommendations for ZK proof systems and is gradually moving toward something like a standard.

National standards bodies in some countries are exploring ZK standards for specific applications like digital identity.

For PQC ZK, formal standardization is years away. In the meantime, the production deployments rely on the academic literature and the security analysis of specific implementations.

For more on standards and FIPS, see [nist-fips-guide](nist-fips-guide.md).

## Privacy Implications and Real-World Examples

Beyond the technical capabilities, post-quantum zero-knowledge has real privacy implications. Several deployed examples illustrate the value:

Privacy-preserving authentication. Some authentication systems use ZK to verify credentials without revealing them. The user proves they hold a valid credential without sending the credential itself. If the underlying ZK is post-quantum secure, the privacy guarantee survives the arrival of quantum computers.

Anonymous credentials. Schemes like Sovrin and Hyperledger AnonCreds let an issuer give a user a credential, and the user later proves possession without revealing the credential or even who they are. PQ versions of these schemes are an active area of work.

Compliant DeFi. Decentralized finance protocols are increasingly using ZK to satisfy regulatory requirements (proof of accreditation, proof of jurisdictional eligibility) without revealing user identities. Long-term durability of these proofs depends on PQ security.

Proof of solvency. Exchanges use ZK to prove they hold customer assets without revealing exact balances. The proofs need to remain credible for years, and PQ ZK is the natural choice for long-term durability.

Each of these use cases benefits from post-quantum security because the proof of compliance, identity, or solvency must remain trustworthy even when quantum computers arrive. Re-proving with new technology is possible but operationally painful, especially when the underlying claim refers to a historical state.

## Open Problems

Several open problems define the research frontier:

Reducing prover time for lattice-based ZK is a major focus. Current systems can take seconds to minutes for moderate computations. Bringing this down to sub-second is the bar for general adoption.

Recursive PQ proofs at production performance. Recursion is theoretically possible but practical recursion in PQ schemes is still being engineered.

Standardization. The lack of formal standards is a barrier to adoption in regulated industries. ZKProof and other groups are working toward documents that can be referenced in compliance frameworks.

Hardware acceleration. As with FHE, dedicated hardware can dramatically speed up PQ ZK proving. Several research projects are pursuing this.

The trajectory is positive. The tools are getting faster, the proof sizes are getting smaller, and the standards are getting more concrete. Five years from now, post-quantum zero-knowledge will likely be a standard part of the cryptographic toolbox.

## Frequently Asked Questions

**Are STARKs post-quantum?**
Yes. STARKs rely only on cryptographic hash functions and finite-field arithmetic, both of which are post-quantum secure.

**Can I use existing zero-knowledge libraries today and migrate later?**
Yes. Most ZK libraries can be replaced with new ones if the application's interface stays consistent. The migration is mostly engineering work rather than cryptographic redesign.

**Are post-quantum ZK proofs slower than classical ones?**
Currently yes, particularly on the prover side. STARK proving is faster than lattice ZK proving but slower than pairing-based SNARKs. Verifier performance is generally acceptable for all approaches.

**What is the size of a typical PQ ZK proof?**
STARK proofs are tens to hundreds of kilobytes. Lattice ZK proofs are kilobytes to tens of kilobytes. Hash-based ZK proofs can be larger still.

**Are blockchain ZK rollups already PQ secure?**
StarkNet and StarkEx (using STARKs) are PQ secure for the proof system. Their other crypto (signatures, hash functions used for state) needs separate migration. zkSync and Polygon zkEVM use SNARKs that are not currently PQ secure.

## Sources

- [NIST Post-Quantum Cryptography Project](https://csrc.nist.gov/projects/post-quantum-cryptography)
- [NIST FIPS 204: ML-DSA Standard](https://csrc.nist.gov/pubs/fips/204/final)
- [NIST FIPS 205: SLH-DSA Standard](https://csrc.nist.gov/pubs/fips/205/final)
- [IETF Crypto Forum Research Group](https://datatracker.ietf.org/rg/cfrg/about/)
- [NIST Internal Report 8413: Status Report on Third Round PQC](https://csrc.nist.gov/pubs/ir/8413/upd1/final)
- [NIST SP 800-208: Stateful Hash-Based Signature Schemes](https://csrc.nist.gov/pubs/sp/800/208/final)

## Related Articles

- [What Is Post-Quantum Cryptography?](what-is-post-quantum-cryptography.md)
- [ML-KEM Explained](ml-kem-explained.md)
- [Why RSA-2048 Will Break](why-rsa-2048-will-break.md)
- [LMS Stateful Signatures](lms-stateful-signatures.md)
- [NIST FIPS Guide](nist-fips-guide.md)

---

### Protect Your Data Before Q-Day Arrives

QNSQY's NIST-standardized post-quantum encryption protects files against both current and quantum-era threats.

[Try QNSQY](../../pricing)
