The Interfold: A Primer on Confidential Coordination 1. Introduction Privacy has become one of the few crypto narratives that has managed to sustain attention this cycle (alongside the recent memecoin <> tokenised stock, perp dexes and prediction markets). The most obvious expression has been Zcash ($ZEC) where it facilitates private transactions and has the “bitcoin but private and quantum proof” narrative as well. With AI becoming increasingly advanced, crypto is becoming much easier to surveil and more sophisticated tools have been built to interpret huge amounts of public on-chain data. This is where $ZEC makes intuitive sense as wealthy participants want to transfer money without broadcasting their entire financial history. But private transactions are only one part of the problem. Consider some of the following situations as well: • People want to vote privately, but the votes still need to produce a public tally • Firms want to participate in a sealed-bid auction without exposing their bids • Banks want to calculate their combined exposure to an asset without revealing their individual book In each case, hiding the data isn't enough. Private information still has to interact with other private information and eventually produce a result everyone can trust. This is the problem @theInterfold is trying to solve. Back in May, @VitalikButerin wrote a detailed post about Interfold and its connection with his MACI (Minimal Anti-Collusion Infrastructure) vision (voter anonymity + verifiable correctness + coercion-resistant voting through an M-of-N committee). He also pointed out the biggest limitation: proving complicated computation over FHE is still expensive. Additive workloads such as vote tallying are much more practical than programs involving lots of multiplication and more complicated manipulation. TLDR: lets competing companies or complete strangers compute together on sensitive data while distributing key generation and decryption across an M-of-N ciphernode committee (private inputs, verifiable outputs, no trusted hardware). The Interfold is effectively taking this architecture beyond voting and applying it to a wider range of confidential coordination problems. 2. What is The Interfold The Interfold is a distributed network for confidential coordination. In a nutshell: private information goes into the network → a predefined computation runs over that info (which remains encrypted) → users encrypt inputs to a shared public key generated by an independent Interfold committee → the committee jointly handles threshold decryption → only the intended result gets decrypted → anyone can verify how that result was produced The project describes this as allowing independent parties to produce shared, verifiable outcomes from private inputs without pooling their data, relying on trusted hardware, or giving one operator unilateral control over the execution. a) The Interfold decentralised network • Encryption can keep data private without necessarily decentralizing control: a single party holding the decryption key could still decide when an output becomes readable, while input validation and computation introduce separate trust assumptions. • The Interfold instead distributes that authority across a temporary committee of operators that never sees the raw private data but collectively manages the cryptographic process and decryption • TLDR: Ethereum decentralizes who validates shared state, while Interfold decentralizes who can unlock private outputs and makes the computation producing those outputs verifiable. b) Ciphernodes • Ciphernodes are Interfold’s cryptographic operators. For each confidential computation, a temporary committee is selected to generate the shared encryption key, and jointly decrypt the approved result. They do not run the application computation itself, a separate Compute Provider runs the FHE program • No single ciphernode holds the full decryption key. Through Distributed Key Generation, the committee creates a shared public key while each operator holds only a cryptographic share • Once computation finishes, no single ciphernode can reveal the result on its own. Enough committee members must contribute valid decryption pieces before the output can be recovered, meaning compromising 1 operator alone is not enough to break confidentiality. The cryptographic process can also be verified, while ethereum:0xe172e9b6cfbeeb5593bdce3f077356fdb33af904 bonding adds an economic cost to malicious behaviour. • This is where ethereum:0xe172e9b6cfbeeb5593bdce3f077356fdb33af904 comes in, it provides the economic security layer. Ciphernode operators bond ethereum:0xe172e9b6cfbeeb5593bdce3f077356fdb33af904 to participate, earn fees/incentives for correct participation, and can be penalized for provable misconduct c) E3s: where the encrypted computation actually happens • E3 (Encrypted Execution Environment) is a temporary environment for specific confidential computation (such as tallying a vote, clearing an auction) with predefined rules for the program • Once the encrypted inputs are processed and verified, the E3 closes d) Use cases • Private governance & markets: secret-ballot voting with private, receipt-free votes and publicly verifiable results, plus sealed-bid auctions where bids stay hidden while only the winner/clearing result is revealed • CRISP (Coercion-Resistant Impartial Selection Protocol): lets people cast private on-chain votes that can still be publicly verified and tallied, without trusting a single party to see or control the ballots. @AragonProject is Interfold’s first live partner where they shipped CRISP as an Aragon plugin so DAOs can run verifiable secret ballots then execute the outcome on-chain • Private data & AI coordination: multiple institutions or agents can combine sensitive data, scores or signals to produce shared statistics, risk metrics, rankings or model evaluations without exposing the underlying inputs This is also where the comparison with TEEs gets interesting 3. TEEs vs E3s Most confidential-computing systems today use TEEs (Trusted Execution Environments). A TEE basically creates a protected hardware environment in which sensitive data can be decrypted and computation can happen fast, but importantly away from the rest of the host system. E3s take a different approach - the private inputs remain encrypted while computation takes place, while the decryption authority is spread across the ciphernode committee. So generally higher computational overhead (higher cryptographic cost), but offers a different privacy and verification model. 4. The ethereum:0xe172e9b6cfbeeb5593bdce3f077356fdb33af904 token ethereum:0xe172e9b6cfbeeb5593bdce3f077356fdb33af904 currently has 3 primary roles • Ciphernode bonding: Operators bond FOLD to become eligible to participate in the network. As more economically valuable computations rely on the network, the amount of economic security behind the operator set becomes increasingly important • Operator incentives: Ciphernodes that successfully complete E3 duties receive fees and ethereum:0xe172e9b6cfbeeb5593bdce3f077356fdb33af904 incentives (bootstrapped the operator network) • Governance: locked to participate in Interfold governance Important to note that E3 usage fees are paid with USDC, So ethereum:0xe172e9b6cfbeeb5593bdce3f077356fdb33af904 isn't currently a straightforward gas token where every computation creates spot demand. The usage also doesn't require the protocol buying or burning FOLD buying or burning. Instead, we think value capture looks something like this: more E3 demand → more fees to operators who will have more inceptives to run the ciphernotes (bond and lock more ethereum:0xe172e9b6cfbeeb5593bdce3f077356fdb33af904) → larger network can can support increasingly valuable applications → leads to more E3 demand. Useful comparisons for FOLD iinclude ethereum:0xa12cc123ba206d4031d1c7f6223d1c2ec249f4f3 and $ZEC • At current time of writing, ethereum:0xa12cc123ba206d4031d1c7f6223d1c2ec249f4f3 has a $110m market cap and $550m FDV • While $ZEC is at a $13b+ market cap / FDV 5. Considerations and things to take note of • Token value capture for ethereum:0xe172e9b6cfbeeb5593bdce3f077356fdb33af904 is more indirect than the transaction-fee buyback and burn model, it is more so a bet on the network demand to increase the economic importance of operators and bonded FOLD • Token price has come down after the team postponed the live demo with @ethereumfndn to October, but many forget it is still an @Official_Upbit listed token at a $23m maketcap • ethereum:0xe172e9b6cfbeeb5593bdce3f077356fdb33af904 also doesn't have $ZEC's private money narrative which is easy for people to understand • FHE performance remains a constraint. The more complex Interfold's applications become, the harder the encrypted computation and verification problem gets • The biggest question is commercial demand - who really needs this tech enough to pay for it repeatedly? There are early live use cases, but the bigger question is whether that translates into recurring demand at scale Interfold's answer to private info is to split authority across temporary cryptographic committees, perform the computation on encrypted data, then reveal only the result. CRISP gives it a credible first application because secret-ballot voting has a need for exactly these properties + Vitalik sees the project as a generalized implementation of ideas he has been pushing since MACI. From here, the opportunity is whether the same architecture can benefit auctions, institutional data and eventually AI coordination. There is already an early partnership through Aragon, with CRISP integrated for verifiable secret-ballot governance. The bigger question is whether demand goes beyond just governance into other recurring use cases, and whether “confidential coordination” becomes a meaningful piece of crypto infra that ethereum:0xe172e9b6cfbeeb5593bdce3f077356fdb33af904 can capture economic value from. I have a position in ethereum:0xe172e9b6cfbeeb5593bdce3f077356fdb33af904