Voting in crypto governance has a dirty secret: it’s not actually secret. Most DAO votes are cast from pseudonymous wallets on public blockchains, meaning anyone with a block explorer can see exactly how you voted. That creates a problem that goes beyond mere privacy discomfort. It opens the door to coercion, vote buying, and social pressure that undermines the entire point of democratic decision-making.
CRISP, short for Coercion-Resistant Impartial Selection Protocol, is a new attempt to fix that. Launched in May 2026 by the Interfold project, which evolved from Gnosis Guild’s Enclave, the protocol brings together three heavyweight cryptographic techniques to create what amounts to a digital secret ballot.
How CRISP actually works
The protocol rests on three cryptographic pillars: fully homomorphic encryption (FHE), zero-knowledge proofs (ZKPs), and distributed threshold cryptography (DTC). Each solves a different piece of the private voting puzzle.
Fully homomorphic encryption is the star of the show. In English: it lets you perform math on encrypted data without ever decrypting it. Think of it like a sealed ballot box that can count its own contents without anyone opening it. Votes go in encrypted, get tallied while still encrypted, and only the final result gets revealed.
Zero-knowledge proofs handle the verification side. They let the system confirm that a vote is valid, that the voter is eligible, and that the final tally is correct, all without revealing any individual vote.
Distributed threshold cryptography is the decentralization layer. Instead of trusting a single party to hold the decryption key, CRISP splits that responsibility across a network of economically incentivized node operators called Ciphernodes. No single Ciphernode can decrypt anything on its own. A threshold number of them must cooperate to reveal the final result.
The combination addresses several known failure modes in existing digital voting. Commit-reveal schemes, which ask voters to commit a hidden vote and reveal it later, are vulnerable to manipulation during the reveal phase. Trusted operator models put too much power in one set of hands. And plain on-chain voting, as mentioned, is about as private as a glass house.
What makes it coercion-resistant
CRISP is designed to be receipt-free, meaning voters cannot generate proof of how they voted even if they wanted to. This breaks the coercion chain entirely. You can promise someone you’ll vote however they like, take their money, and vote however you actually want. There’s no receipt to check.
The protocol also supports censorship-resistant submission, meaning votes can’t be selectively blocked or filtered by any intermediary. Combined with anonymous participant engagement, the system aims to make the act of voting both private and unstoppable.
A live proof-of-concept demo is currently available at crisp.enclave.gg for anyone who wants to see the mechanism in action rather than take it on faith.
Open source, no token, broader ambitions
One notable feature of the project is what it doesn’t have: a token. There’s no native cryptocurrency, no market listing, no liquidity pool. The entire codebase is open source, hosted on GitHub under gnosisguild/enclave.
The Interfold project, which rebranded from the Enclave framework, positions itself as infrastructure rather than a financial product. The team frames their work around encrypted execution environments, or E3s, a broader concept that extends beyond voting into any scenario where computation needs to happen on sensitive data.
Gnosis Guild applied for a Zcash Community Grant in March 2026 to develop what they’re calling “Zecret Ballots,” an integration that would bring CRISP’s coercion-resistant voting capabilities specifically to the Zcash community.