The Battle for Blockchain Privacy
In the previous article I looked at how zero-knowledge proofs are solving blockchain's privacy paradox. But not all ZK-proofs are created equal. Two technologies are competing to become the standard for Ethereum's privacy future: zkSNARKs and zkSTARKs.
It is a bit like VHS against Betamax in the 1980s. Both work, and each has distinct advantages that could decide the winner.
zkSNARK: The Compact Powerhouse
zkSNARK stands for Zero-Knowledge Succinct Non-Interactive Argument of Knowledge.
Key Characteristics
The defining trait is size. A SNARK proof is typically around 200 bytes, which makes on-chain verification cheap. Small proofs also verify quickly, and quick verification is what keeps gas costs down on Ethereum. Their data availability is better too, which is a big part of why they are the more popular choice in L1 rollups. And they are battle-tested: the technology is mature and has years of production use behind it.
The Achilles heel is the trusted setup.
zkSNARKs need a "trusted setup" ceremony where the initial parameters are generated. If those parameters are compromised, the whole security model breaks down. Think of it as cutting a master key. If someone keeps a copy, they can forge proofs and nobody will ever know.
In practice SNARKs already run a lot of real systems. Polygon Hermez and zkSync Era use them for blockchain scaling, Zcash and Tornado Cash use them for private transactions, and the zk-creds protocol uses them to protect identity through anonymous credentials.
zkSTARK: The Transparent Challenger
zkSTARK stands for Zero-Knowledge Scalable Transparent Argument of Knowledge.
Key Characteristics
STARKs remove the weak spot SNARKs carry around. There is no trusted setup at all: they use public randomness and hash functions, so there is no secret parameter for anyone to keep a copy of. The same hash-based construction makes them resistant to future quantum computer attacks. They also perform better as the computational load grows, and the proof generation process is fully transparent.
The trade-off is size.
A STARK proof can be 10 to 100 times larger than a SNARK. That means higher on-chain storage costs, longer verification times and more bandwidth to move the proof around.
Pioneer Project - Starkware
Starkware is the company pushing zkSTARKs hardest. It built Cairo, a ZK-friendly programming language designed specifically for STARKs. It runs Starknet, a layer-2 network processing millions of transactions with provable integrity. And its StarkEx engine powers dYdX, Immutable X and other major platforms.
The Technical Deep Dive
Proof Size Comparison
| Technology | Proof Size |
|---|---|
| zkSNARK | ~200 bytes |
| zkSTARK | ~2,000-20,000 bytes (10-100x larger) |
Verification Time
A zkSNARK verifies in constant time, and that time is very short. A zkSTARK's verification time grows logarithmically with the size of the computation, so it gets slower as the work gets bigger, but slowly.
Security Model
The two rest on different cryptography. zkSNARKs rely on elliptic curve cryptography, which is vulnerable to quantum attacks. zkSTARKs are built on hash functions, which are quantum-resistant.
Who's Winning?
Current Market Share
Right now zkSNARKs dominate the rollup space because they are cheaper to run. zkSTARKs are gaining ground in applications where transparency matters most.
The Verdict
I do not think this is about which one "wins". It is about which tool fits your use case.
Choose a zkSNARK when you need a minimal on-chain footprint, the fastest possible verification, compatibility with existing Ethereum infrastructure and lower costs today. Choose a zkSTARK when you care more about having no trusted setup (maximum decentralization), quantum resistance as future-proofing, transparency in the proof process, or scaling to massive computation.
The Hybrid Future
The interesting part is that the industry is moving toward using both. Ethereum is exploring each technology at layer 1 for different use cases. Recursive proofs combine SNARKs and STARKs to get the best of both. And DeFi protocols are picking one or the other based on their own security and cost requirements, application by application.
What This Means for Ethereum
As Ethereum pushes toward integrating ZK-proofs at the base layer, the competition between SNARKs and STARKs is driving innovation fast. Gas costs for ZK verification are dropping by 90% or more, proving times are falling from hours to minutes, and new programming languages are making ZK development accessible to more developers.
The real winner is users, who will soon get privacy and scalability without having to give up one for the other.
Next in this series: Ethereum's official 2025 Privacy Roadmap, and how the Ethereum Foundation is making privacy a core feature, not an afterthought.
Which technology do you think will dominate? Are you team SNARK or team STARK?
Keep reading
- Zero-Knowledge Proofs Explained: The Cryptography Behind RAILGUN
We've covered what RAILGUN does and how to use it. Now let's explore the magic underneath: zero-knowledge proofs. This is the cryptographic foundation that makes private transactions possible without requiring trust in any central party.
- zkTLS: Bridging Web2 Privacy with Web3 Verification
I've built several dApps and AI agents using Zero-Knowledge Transport Layer Security (zkTLS) and wanted to share why this technology is so transformative. zkTLS represents a significant advancement in blockchain security architecture, combining zero-knowledge cryptography with enhanced TLS protocols.
- Blockchain's Privacy Paradox and the Solution: Zero-Knowledge Proofs
For the past 10 years, blockchain has operated with radical transparency - every transaction is publicly recorded, which leaves users vulnerable to phishing attacks. While transparency is blockchain's greatest strength for trust and verification, it's also becoming its Achilles heel for mainstream adoption.