A zero-knowledge proof lets one party (the prover) convince another party (the verifier) that a statement is true without revealing any information beyond the fact that it is true. You prove you know a secret without revealing the secret. In crypto, ZK proofs have two major application categories: scaling (ZK-rollups compress many transactions into a single proof that Ethereum verifies cheaply) and privacy (ZK proofs let users prove eligibility or identity without revealing personal data).
How zero-knowledge proofs work
A ZK proof has three properties. Completeness: if the statement is true, an honest prover can always convince the verifier. Soundness: if the statement is false, a cheating prover can only convince the verifier with negligible probability. Zero-knowledge: the verifier learns nothing beyond the truth of the statement itself. Practical ZK proof systems (SNARKs and STARKs) convert computational statements into polynomial equations and then prove knowledge of the solution without revealing the inputs.
SNARKs (Succinct Non-Interactive Arguments of Knowledge) require a trusted setup ceremony to generate public parameters; anyone who participated in the ceremony could theoretically compromise the system if they colluded. STARKs (Scalable Transparent Arguments of Knowledge) require no trusted setup and are post-quantum secure, but produce larger proofs. Both are used in production: Groth16 SNARKs in Zcash, STARKs in StarkNet, and PLONK (a universal SNARK without per-circuit setup) in many ZK-rollups including zkSync and Polygon zkEVM.
What this means for traders
ZK proofs in scaling are already live and affecting trading economics. The fee reduction from ZK-rollups (see ZK-rollups explained and EIP-4844) directly reduces costs for every on-chain trade. The proving time improvement between 2022 and 2026 was roughly 100x: what took 10 minutes to prove in 2022 takes 6 seconds in 2026 for equivalent complexity, due to hardware acceleration (ASICs, FPGAs) and algorithmic improvements.
ZK for privacy and identity is an emerging category. Zcash uses ZK proofs for private transactions: you prove you are authorized to spend funds without revealing which funds you hold. Proof of solvency for exchanges uses ZK proofs to prove liabilities without revealing individual user balances. ZK-KYC (proving you are a verified adult or non-sanctioned individual to access a protocol without revealing your passport number) is in production in limited jurisdictions and could change how regulated DeFi handles compliance without on-chain identity exposure.
A concrete example
You want to prove you are a US non-sanctioned individual to access a compliant DeFi protocol without revealing your name or passport to the protocol. A ZK-KYC system: a trusted KYC provider (Coinbase, Binance, a licensed identity provider) verifies your identity off-chain. They issue a cryptographic credential. You generate a ZK proof that you hold a valid credential from an approved provider, without revealing which provider or what your identity is. The protocol accepts the proof, grants access, and has no record of who you are beyond the fact that someone with a valid credential interacted. As of 2026, Polygon ID and similar systems are deployed in limited production contexts for this use case.
Frequently asked questions
What is a trusted setup and why does it matter?
A trusted setup ceremony generates the public parameters (structured reference string) that a SNARK proof system uses. The ceremony works by having multiple participants each contribute random numbers; the parameters are computed from these contributions. If every single participant destroys their contribution after the ceremony, the parameters are safe. If even one participant kept their contribution secret, the system is secure. Powers of Tau, Zcash’s Sapling ceremony, and Ethereum’s KZG ceremony all used this multi-party approach with thousands of contributors to minimize the trust assumption.
What is recursive proof composition?
Recursive ZK proofs allow a proof to verify another proof. This enables aggregation: instead of verifying 1,000 individual proofs on Ethereum, you generate one proof that verifies all 1,000 previous proofs, and Ethereum verifies only that final proof. This is how ZK-rollups scale beyond their initial throughput limits: recursive proofs compress arbitrarily many transactions into a single fixed-size verification. StarkNet and Polygon use recursive proving to stack transaction batches before settling on Ethereum.
Can ZK proofs be broken by quantum computers?
SNARKs based on elliptic curve pairings (Groth16, PLONK) would be vulnerable to a sufficiently powerful quantum computer using Shor’s algorithm. STARKs, which rely on hash functions rather than elliptic curves, are considered post-quantum secure. This is one reason Ethereum’s long-term roadmap considers STARKs for future applications, even though the immediate quantum threat to live systems is minimal given current hardware limitations.
