Zero-Knowledge Proofs: Why the Cryptography Behind Privacy Is Becoming Infrastructure

The core concept sounds like mathematical magic: prove you know something without revealing what you know. In practice, *zero-knowledge proofs* are a decades-old cryptographic primitive that spent most of its life in academic papers — until recently.

The ZK proof is now infrastructure.

## The Basic Mechanics

A ZK proof allows a *prover* to convince a *verifier* that a statement is true without revealing any information beyond the truth of the statement itself. The classic example: prove you know the password to a door without saying the password.

In 1985, Goldwasser, Micali, and Rackoff published the foundational paper defining interactive zero-knowledge proofs. For three decades, ZK proofs were computationally impractical at scale — the math worked but the hardware couldn't run it in reasonable time. What changed is proving efficiency: a combination of better algorithms and faster hardware reduced proving time from hours to seconds.

## The Pivot: From Zcash Privacy to zkEVM Scaling

*Zcash*, launched in 2016, was the first major production deployment of ZK proofs in cryptocurrency — specifically zk-SNARKs (*Succinct Non-Interactive Arguments of Knowledge*) for shielded transactions. The idea was compelling: transact on a public blockchain with cryptographic privacy guarantees. Zcash never achieved mainstream adoption, but it stress-tested ZK proofs in production at real transaction volumes.

The pivot came when Ethereum scaling projects realized ZK proofs could do something entirely different: *verify computation*. Instead of proving "I know a secret," a zkEVM proves "this batch of Ethereum transactions was executed correctly according to the EVM spec." That proof is then verified on Ethereum's mainnet in milliseconds at minimal cost. The scalability implications are significant.

## Proving Systems: Why Groth16 vs PLONK vs STARKs Matters

The choice of proving system determines trust assumptions, setup requirements, and performance characteristics — and these differences matter in production.

**Groth16** (the original zk-SNARK variant) is fast and generates small proofs, but requires a *trusted setup* — a ceremony where participants collectively generate cryptographic parameters. If participants collude, the proof system breaks. Zcash ran the "Powers of Tau" ceremony in 2017; Groth16 circuit-specific setups require a new ceremony for each new circuit.

**PLONK** uses a *universal* trusted setup: one ceremony covers many different applications. More flexible than Groth16, somewhat larger proofs, but the universal setup dramatically reduces operational overhead.

**STARKs** (*Scalable Transparent Arguments of Knowledge*) require no trusted setup and are quantum-resistant — their security relies on hash functions rather than elliptic curve assumptions. The trade-off: significantly larger proofs, which increase on-chain verification costs. StarkWare built STARKs for StarkNet and their StarkEx deployments.

The reason this matters practically: systems with trusted setups have a specific failure mode that transparent systems don't. The choice shapes both security properties and operational complexity for anything built on top.

## Current Deployment

The zkEVM landscape in 2025 includes zkSync Era (Matter Labs), StarkNet (StarkWare), Polygon zkEVM, and Scroll — each making different proving system trade-offs. These aren't research projects. zkSync Era processed over $10 billion in transaction value in 2024. Polygon zkEVM handles significant DeFi volume daily. Proving time for a batch of transactions has dropped from hours to under 10 seconds over four years of engineering.

## Beyond Crypto

ZK proofs are transitioning from privacy technology to general-purpose verifiable computation — and the use cases go far beyond blockchain. This is the development that most coverage misses.

**Identity verification**: prove you're over 18 without revealing your birthdate. Prove you pass an AML check without sharing your financial history. Several EU identity projects are exploring ZK-based credential systems.

**Private data computation**: prove a machine learning model produced a specific output without revealing training data. Relevant for healthcare models running on sensitive patient data.

**Compliance reporting**: prove regulatory compliance without disclosing proprietary business data. Relevant for financial institutions reporting to multiple jurisdictions.

ZK proofs are becoming what cryptographic signatures became in the 1990s — infrastructure that enables digital systems to be more trustworthy without requiring parties to fully trust each other.

> **Key Takeaway:** ZK proofs have moved from academic curiosity to production infrastructure, processing billions in transaction volume across multiple networks. The technical differences between proving systems matter for trust and performance. More significantly, the application space is expanding beyond blockchain privacy into general-purpose verifiable computation — identity, compliance, and private AI are the next wave.

Zero-Knowledge Proofs: Why the Cryptography Behind Privacy Is Becoming Infrastructure

// COMMENTS

ON THIS PAGE