Zama 백서

bitpro가 원문에서 글자만 모은 내용이에요. 표·그림·수식은 빠지거나 깨질 수 있어요. 전체 75쪽 중 2쪽

sability — encrypted data can be shared between contracts and chains. * The private FHE key is never directly accessible by a single party; instead, it is secret-shared among the MPC nodes. The system follows the NIST SP 800-57 key lifecycle model, managing key states such as Active, Suspended, Deactivated,and Destroyed to ensure proper rotation and forward security. ### Threshold Decryption via MPC The KMS performs decryption using a threshold decryption protocol — at least a minimum number of MPC parties (e.g., 9 out of 13) must participate in the protocol to robustly decrypt a value. * This protects against compromise: no individual party has access to the full key. And adversary would need to control more than the threshold of KMS nodes to influence the system. * The protocol supports both: * Public decryption (e.g., for smart contracts) * User decryption (privately returned, re-encrypted only for the user to access) All decryption operation outputs are signed by each node and the output can be verified on-chain for full auditability. ### ZK Proof support The KMS generates Common Reference Strings (CRS) needed to validate Zero-Knowledge Proofs of Knowledge (ZKPoK) when users submit encrypted values. This ensures encrypted inputs are valid and well-formed, and that a user has knowledge of the plaintext contained in the submitted input ciphertext. ## Security architecture ### MPC-based key sharing * The KMS currently uses 13 MPC nodes, operated by different reputable organizations. * Private keys are split using threshold secret sharing. * Communication between nodes are secured using mTLS with gRPC. ### Honest majority assumption * The protocol is robust against malicious actors as long as at most 1/3 of the nodes act maliciously. * It supports guaranteed output delivery even if some nodes are offline or misbehaving. ### Secure execution environments Each MPC node runs by default inside an AWS Nitro Enclave, a secure execution environment that prevents even node operators from accessing their own key shares. This design mitigates insider risks, such as unauthorized key reconstruction or selling of shares. ### Auditable via gateway * All operations are broadcast through the Gateway and recorded as blockchain events. * KMS responses are signed, allowing smart contracts and users to verify results cryptographically. ### Key lifecycle management The KMS adheres to a formal key lifecycle, as per NIST SP 800-57: | State | Description | | -------------- | ------------------------------------------------------------------ | | Pre-activation | Key is created but not in use. | | Active | Key is used for encryption and decryption. | | Suspended | Temporarily replaced during rotation. Still usable for decryption. | | Deactivated | Archived; only used for decryption. | | Compromised | Flagged for misuse; only decryption allowed. | | Destroyed | Key material is deleted permanently. | The KMS supports key switching using FHE, allowing ciphertexts to be securely transferred between keys during rotation. This maintains interoperability across key updates. ### Backup & recovery In addition to robustness through MPC, the KMS also offers a custodial backup system: * Each MPC node splits its key share into encrypted fragments, distributing them to independent custodians. * If a share is lost, a quorum of custodians can collaboratively restore it, ensuring recovery even if several MPC nodes are offline. * This approach guarantees business continuity and resilience against outages. * All recovery operations require a quorum of operators and are fully auditable on-chain. ### Workflow example: Public decryption 1. A smart contract requests decryption via an oracle. 2. The Gateway verifies permissions (i.e. that the contract is allowed to decrypt the ciphertext) and emits an event. 3. KMS parties retrieve the ciphertext, verify it, and run the MPC decryption protocol to jointly compute the plaintext and sign their result. 4. Once a quorum agrees on the plaintext result, it is published (with signatures). 5. The oracle posts the plaintext back on-chain and contracts can verify the authenticity using the KMS signatures. # Roadmap This document gives a preview of the upcoming features of FHEVM. In addition to what's listed here, you can [submit your feature request](https://github.com/zama-ai/fhevm/issues/new) on GitHub. ## Features | Name | Description | ETA | | ---------------- | --------------------------------------------------------- | ----------- | | Foundry template | [Forge](https://book.getfoundry.sh/reference/forge/forge) | Coming soon | ## Operations | Name | Function name | Type | ETA | | --------------------- | ----------------- | ------------------ | ----------- | | Signed Integers | `eintX` | | Coming soon | | Add w/ overflow check | `FHE.safeAdd` | Binary, Decryption | Coming soon | | Sub w/ overflow check | `FHE.safeSub` | Binary, Decryption | Coming soon | | Mul w/ overflow check | `FHE.safeMul` | Binary, Decryption | Coming soon | | Random signed int | `FHE.randEintX()` | Random | - | | Div | `FHE.div` | Binary | - | | Rem | `FHE.rem` | Binary | - | | Set inclusion | `FHE.isIn()` | Binary | - | {% hint style="info" %} Random encrypted integers that are generated fully on-chain. Currently, implemented as a mockup by using a PRNG in the plain. Not for use in production! {% endhint %} # Contributing There are two ways to contribute to FHEVM: * [Open issues](https://github.com/zama-ai/fhevm/issues/new/choose) to report bugs and typos, or to suggest new ideas * Request to become an official contributor by emailing <hello@zama.ai>. Becoming an approved contributor involves signing our Contributor License Agreement (CLA). Only approved contributors can send pull requests, so please make sure to get in touch before you do! ## Zama Bounty Program Solve challenges and earn rewards: * [bounty-program](https://github.com/zama-ai/bounty-program) - Zama's FHE Bounty Program # Zama Confidential Blockchain Protocol Litepaper {% hint style="info" %} This litepaper describes Zama’s Confidential Blockchain Protocol, which enables confidential smart contracts on existing public blockchains. It includes details about the protocol and token, as well as documentation for node operators. \ \ <i class="fa-link">:link:</i> To read Zama's FHEVM technical white paper, please see [on Github](https://github.com/zama-ai/fhevm/blob/main/fhevm-whitepaper.pdf).\ <i class="fa-link">:link:</i> To read the Zama Token White Paper, see [this page](https://docs.zama.org/protocol/mica). {% endhint %} ## The blockchain confidentiality dilemma Why do we need blockchain? This question often comes along when discussing building decentralized applications (dapps). After all, most things we use today are not blockchain based and work just fine. However, there are some applications where the cost of blindly trusting a third party and being wrong is too high, such as when dealing with financial assets, identity or governance. In those cases, consumers and companies want strong guarantees that whatever service is being provided is done correctly, while service providers want to ensure their users have the right to use the assets/data they claim. Blockchains solve this by enabling anyone to publicly verify that a request was executed according to a predetermined logic, and that the resulting state of the overall system is correct. Service providers and their customers no longer have to trust each other, as the integrity of the transaction is guaranteed by the blockchain itself. One major issue with public verifiability however is that it requires disclosing all the transactions and data to everyone, as keeping them private would prevent verifiability in the first place. This lack of confidentiality has been a major hindrance to global adoption of blockchains, as the very data it is supposed to be used for (money, identity, …) is highly sensitive by nature. Without confidentiality, blockchain cannot reach mass adoption. ## The Zama Confidential Blockchain Protocol The Zama Confidential Blockchain Protocol (or simply the Zama Protocol) enables issuing, managing and trading assets confidentially on existing public blockchains. It is the most advanced confidentiality protocol to date, offering: * **End-to-end encryption** of transaction inputs and state: no-one can see the data, not even node operators. * **Composability** between confidential contracts, as well as with non-confidential ones. Developers can build on top of other contracts, tokens and dapps. * **Programmable confidentiality**: smart contracts define who can decrypt what, meaning developers have full control over confidentiality rules in their applications. The Zama Protocol is not a new L1 or L2, but rather a cross-chain confidentiality layer sitting on top of existing chains. As such, users don’t need to bridge to a new chain and can interact with confidential dapps from wherever they choose. It leverages Zama’s state-of-the-art Fully Homomorphic Encryption (FHE) technology, which enables computing directly on encrypted data. FHE has long been considered the “holy grail” of cryptography, as it allows end-to-end encryption for any application, onchain or offchain. We believe that just like the internet went from zero encryption with HTTP to encrypting data in transit with HTTPS, the next natural step will be to use FHE to enable end-to-end encryption by default in every application, something we call [HTTPZ](https://www.zama.ai/post/people-should-not-care-about-privacy). Until recently however, FHE was too slow, too limited in terms of applications it could support, and too difficult to use for developers. This is what our team at Zama has spent the last 5 years solving. We now have a highly efficient FHE technology that can support any type of application, using common programming languages such as Solidity and Python, while being over 100x faster than 5 years ago. Importantly, Zama’s FHE technology is already post-quantum, meaning there is no known quantum algorithms that can break it. While FHE is the core technology used in the Zama Protocol, we also leverage Multi-Party Computation (MPC) and Zero-Knowledge Proofs (ZK) to address the shortcomings of other confidentiality solutions: * FHE enables confidentiality while being fully publicly verifiable (anyone can recompute the FHE operations and verify them). Using GPUs will soon allow scaling to 100+ transactions/s while dedicated hardware accelerators (FPGAs and ASICs) will enable scaling to thousands of transactions per second. * MPC enables decentralizing the global network key, ensuring no single party can access it. Using MPC only to generate keys and decrypt data for users minimizes latency and communication, thereby making it far more scalable and decentralized than using it for private computation. * ZK ensures the encrypted inputs provided by users were actually encrypted correctly. Using ZK only for this specific purpose makes the ZK proofs lightweight and cheap to generate in a browser or mobile app. The table below summarizes the advantages of the Zama Protocol versus other technologies used in confidential blockchain protocols: <table><thead><tr><th width="145.66796875"></th><th align="center">Zama</th><th width="120.30859375" align="center">Other FHE</th><th width="120.66015625" align="center">MPC</th><th width="114.15625" align="center">ZK</th><th align="center">TEE</th><th>Private Chains</th></tr></thead><tbody><tr><td><strong>Secure</strong></td><td align="center">✅</td><td align="center">✅</td><td align="center">✅</td><td align="center">✅</td><td align="center">❌</td><td>✅</td></tr><tr><td><strong>Decentralized</strong></td><td align="center">✅</td><td align="center">✅</td><td align="center">✅</td><td align="center">✅</td><td align="center">✅</td><td>❌</td></tr><tr><td><strong>Verifiable</strong></td><td align="center">✅</td><td align="center">✅</td><td align="center">❌</td><td align="center">✅</td><td align="center">❌</td><td>❌</td></tr><tr><td><strong>Composable</strong></td><td align="center">✅</td><td align="center">✅</td><td align="center">✅</td><td align="center">❌</td><td align="center">✅</td><td>✅</td></tr><tr><td><strong>Scalable</strong></td><td align="center">✅</td><td align="center">❌</td><td align="center">✅</td><td align="center">✅</td><td align="center">✅</td><td>✅</td></tr><tr><td><strong>Easy to use</strong></td><td align="center">✅</td><td align="center">❌</td><td align="center">❌</td><td align="center">❌</td><td align="center">✅</td><td>✅</td></tr></tbody></table> ## Roadmap The Zama Protocol leverages years of research and development work done at Zama. The testnet is already live, with the mainnet on Ethereum launching by end of 2025. The timeline is as follows: * **Ethereum Mainnet (live)**. This will be the first official mainnet bringing confidentiality to Ethereum.&#x20; * **Other EVM chains (H1 2026)**. We will add more EVM chains to the Zama protocol to enable cross-chain confidential assets and applications. * **Solana support (H2 2026)**. After an initial phase of EVM-only support, we will deploy the Zama Protocol on Solana, enabling confidential SVM applications. ## Use cases Confidential smart contracts enable a new design space for blockchain applications, in particular when applied to finance, identity and governance. If we look at web2, it is clear that most applications do not share all the data publicly, and thus it is likely that the vast majority of blockchain applications are yet to be built, now that confidentiality is no longer an issue. Here are some example use cases: ### Finance * **Confidential payments.** Stablecoins are one the most successful use case for blockchain, with trillions in yearly volume. Everything from credit card payments to salaries, remittances and banking rails is now moving onchain. One of the absolute key requirement however is confidentiality and compliance. Thanks to FHE and the Zama Protocol, this is now possible: balances and transfer amounts are kept encrypted end-to-end, while payment providers can embed compliance features into the token contract directly. You can read more about confidential, compliant payments [here](https://www.zama.ai/post/programmable-privacy-and-onchain-compliance-using-homomorphic-encryption). * **Tokenization & RWAs**. The tokenization of financial assets is one of the main adoption drivers of blockchain for large institutions. From fund shares to stocks, bonds or derivatives, there is up to $100T of assets that could potentially move onchain. Due to confidentiality and compliance issues however, TradFi institutions have had to rely on private blockchains, making it difficult to ensure interoperability between institutions. With the Zama Protocol, they can now use existing public blockchain such as Ethereum or Solana to tokenize and trade their assets, while keeping their activity and investor identity confidential. They can also ensure KYC/AML checks are done in the smart contracts directly, without revealing sensitive information to others. You can read more about this use case [in the report published by JP Morgan - Kynexis](https://www.jpmorgan.com/kinexys/documents/JPMC-Kinexys-Project-Epic-Whitepaper-2024.pdf), in which they built a proof-of-concept using Zama’s technology. * **Confidential DeFi.** DeFi has redefined finance by allowing anyone to participate and earn yield, but it suffers from two major issues: people don’t like sharing how much they own, and bots front-running transactions makes it expensive for end users to swap assets onchain. FHE can solve both issues by enabling end-to-end encrypted swaps, where the amount and possibly asset is kept private at all times. Some other use cases includes confidential lending, onchain credit scoring, option pricing and more. ### Tokens * **Sealed-bid auctions.** Sell assets such as NFTs or tokens in an onchain sealed-bid auction. Each participant places an encrypted bid onchain. When the auction ends, the highest bidder(s) win the item(s), without revealing any of the bids. Not only does this enable better price discovery, it also prevents bots from stealing the auction by monitoring the mempool. This is a particularly effective method for public token sales. * **Confidential distributions.** Distributing tokens currently requires disclosing publicly how much each address receives. Whether it’s for airdrops, grants, investors or developers, keeping the distributed amounts private is paramount to privacy and security onchain. With FHE, protocols can distribute their token confidentially, run vesting on those encrypted tokens, enable confidential staking and more. ### Identity and Governance * **Composable onchain identity.** Offchain, we use our identities all the time, from buying products online to booking plane tickets. Doing so onchain however would leak sensitive information such as your name, address, social security number and more. With FHE however, you can have a complete Decentralized ID (DID) + Verifiable Credentials (VC) system onchain, where your identity is encrypted while being fully composable with decentralized applications. Just like you can have account abstraction, you can now have identity abstraction. This is also essential for compliance in onchain payments and tokenization, as it can be used by smart contracts to verify claims in a decentralized, private manner. * **Confidential governance.** The idea of onchain voting, whether for DAOs, companies or governments, has been explored for as long as blockchains exist. But having the votes cast publicly onchain can lead to biases, blackmailing, or bribing. With FHE, votes (and numbers of tokens staked) can be kept private, ensuring only the final tally is revealed, and not the individual votes. ### Other examples * **Onchain corporations.** Managing a company onchain would be impossible without the promise of confidentiality. Indeed, information such as the cap table, financials, board votes, customers, and employee registers should not be disclosed publicly. With FHE, all this information could be kept onchain, allowing smart contracts to automate many day-to-day company operations. **‍‍** * **Prediction markets. ‍‍**Prediction markets are based on the wisdom of the crowd concept: the average prediction of a large number of people tend to be close to the correct outcome. However, this only works if participants are not biased by previous predictions. The Zama Protocol solves this by enabling prediction markets where predictions are encrypted until revealed periodically, leading to better precision in outcomes. * **Data marketplaces for AI. ‍‍**AI strives on data. With FHE, users can selectively share and sell their data with companies wishing to train AI models. More than this, models can potentially be trained encrypted, with only the result being decrypted, ensuring that users have a constant stream of revenue for their data vs selling it only once and it being used forever. These are just some examples of what can be done today. We believe that FHE, through Zama’s Protocol, will enable unprecedented liquidity, enabling users and companies to move onchain. With time and scale, it would even become possible to run entire companies, cities or even countries onchain, including their financial and identity infrastructure, elections, currency, taxes, land, car and company registries. Confidential blockchains don’t just enable programmable money: they enable programmable public infrastructure. ## Creating confidential applications Creating a confidential dapp using existing solutions often requires learning a new (niche) programming language, using dedicated (and often limited) developer tools, and mastering advanced cryptographic concepts. The Zama Protocol on the other hand enables developers to create confidential dapps directly in Solidity, w
프로젝트로 돌아가기