Source: Biteye
For Ethereum and the broader blockchain narrative, there are now many excellent teams launching scaling solutions, but scaling is not the only problem that needs to be solved.
The next key function to be achieved is privacy. The privacy track has recently become a hot spot for infrastructure investment in the primary market.
This article will introduce the implementation of two popular privacy chain technology routes, Zero Knowledge Proof and Fully Homomorphic Encryption, and will also introduce related potential projects that can be paid attention to.
First, let’s discuss a question: Does Web3 have any privacy application scenarios?
01Why does Web3 need privacy?
The existing mainstream chains are all public ledgers, and all transactions are conducted on the chain, which means that the status changes containing asset information related to the address or account are open and transparent.
Initially, information transparency was only a by-product of monitoring consensus security. However, as the industry develops, the consensus mechanism has been gradually optimized, improved, and reliable. The transparent public ledger has gradually become a feature that serves technology arbitrage:
Miners can selectively package transactions based on fees, resulting in transactions with lower fees being less likely to be processed, forcing users to increase gas fees. What is more worrying is the front-running and censorship attacks by miners or block producers who monitor public ledgers.
By monitoring the buy orders on the chain and adding its own buy orders before retail buy orders are executed, this has led to huge security issues. In the past year, MEV has successfully extracted nearly $2 billion from the market.
Such a huge and continuous outflow of funds can be regarded as a huge hidden danger in the development of the crypto market.
At the same time, due to the lack of privacy support, users lose data ownership. Both the asset information and transaction information of the address may be monitored and used. This runs counter to the vision of Web3.
Therefore, when the expansion problem is solved, the privacy smart contract chain becomes the next urgent function to be realized.
To realize privacy smart contracts, three technical routes are currently adopted:
1) TEE (Trusted Execution Environment) solutions represented by Secret Network and Oasis Network, which have been launched but have not been very popular;
2) The zkVM solution, which is based on the ZK (zero-knowledge proof) principle that has come into the public eye through Ethereum zk-rollup;
3) FHE (fully homomorphic encryption) solutions, which have only recently entered the market;
TEE technology is the most mature and has a lot of related documents. Interested readers can learn about it on their own or go to the projects mentioned above to experience it firsthand. Therefore, this article will focus on the more topical zkVM and FHE solutions.
02Zero Knowledge Proof
zkEVM and zkVM
Most ZK solutions fall into two camps: those that are built on top of Ethereum (zkEVM) and those that are custom built (zkVM), and thus may choose to build with a different set of underlying tradeoffs and fundamental parameters.
zkEVM is an Ethereum virtual machine-compatible, zero-knowledge proof-friendly virtual machine that guarantees the correctness of programs, operations, inputs, and outputs.
By building on top of the Ethereum blockchain, the zkEVM model incorporates both Ethereum’s strengths and weaknesses.
Since it optimizes compatibility with the Ethereum network, it benefits from Ethereum's large user base and it is easier for developers to develop on top of it (this is because there are a large number of Solidity developers and its infrastructure (including execution clients) is shared).
However, this also means that its ability to incorporate zero-knowledge proofs and other privacy measures is limited to the built-in limitations of the Ethereum network.
The closer you get to fully emulating the Ethereum model with the zkEVM, the more you pay a performance penalty because it takes longer to generate proofs.
Because computation is done on the blockchain, every transaction is completely public and transparent, which is beneficial for some applications, but for others this lack of privacy is unreasonable or unsafe (for example, applications related to sensitive personal financial information).
zkVM is a virtual machine that guarantees security and verifiable trust through zero-knowledge proofs - you input the old state and program, and it returns the new state in a trusted way. It can optimize the environment and make it cheaper, more efficient, and even easier to integrate zero-knowledge proofs into on-chain transactions.
Essentially, a proper zkVM allows all of its applications to use zero-knowledge proofs with relative ease in every transaction. A true zkVM is one that is built with ZK-first principles in mind and integrates them into every part of the stack.
Ethereum is a completely open and transparent blockchain. If developers try to introduce privacy now, its performance will definitely not be as good as a blockchain that supports privacy from the beginning.
This is difficult from an engineering perspective because developers must code programs that were not designed to run on this type of field, resulting in large and more complex circuits.
Therefore, the performance of zkVM will be better than that of zkEVM, and it is a technical solution that is very worth ambush.
At present, some zkVM solutions have emerged, such as L1: Aleo, Mina, etc.; L2: Aztec, etc. The market expectations of these projects are relatively high, and the cost-effectiveness of participation is not high. The following is a zkVM project that is more suitable for ambush.
Ola Network
Ola is a scalable privacy-preserving and compliance-optimized ZKVM Rollup platform, featuring programmable privacy, scalability, and multi-language compatibility. Ola aims to be a universal Layer2 scaling solution that can add privacy protection and scalability to various programmable Layer1 blockchains.
Ola recently raised $3 million in a seed round led by Web3 Ventures and Foresight Ventures, with participation from Token Metrics Ventures, J17 Capital, Skyland Ventures, LD Capital, and CatcherVC.
Ola's main products include the ZK-optimized virtual machine Ola-VM and the smart contract language Ola-lang.
Ola-lang is a general-purpose language developed based on ZK-VM with higher programmability. Developers can use Ola-lang to flexibly deploy any type of smart contract, whether it is on a public chain or an enterprise-level private chain.
The ZK-optimized virtual machine Ola-VM uses a reduced instruction set architecture and achieves better performance through full ZK support and non-deterministic computing.
Simply put, Ola is building a Layer 2 infrastructure that combines optional privacy and programmability.
It allows the public chain to inherit network security while obtaining functions such as privacy protection and performance expansion by deploying corresponding verification contracts.
This approach avoids sacrificing the programmability and decentralization of the public chain. Developers can add privacy and expansion solutions to different public chains as needed without making any changes on the chain.
This provides customizable privacy and scalability while maintaining the open nature of the public chain.
Currently, Ola has launched tasks in the Ola Gala, which can qualify for the 2024 Ola Public Testnet and receive rewards such as NFTs.
In addition, on November 10, Ola’s official website opened the Devnet test network application. Developers may wish to pay attention to this application. Selected candidates can receive rewards, technical assistance, developer resources, and opportunities to deploy Dapps on the Ola mainnet.
03Fully Homomorphic Encryption
Fully homomorphic encryption is a new technology applied to blockchain. It is one of the public chain solutions that institutions are more interested in after the ZK craze. As a new concept, there are relatively few projects at present, and they are all in the early stages, so it is worth ambush.
Fully homomorphic encryption is an open problem that has been raised in the cryptography community a long time ago. As early as 1978, Rivest, Adleman, and Dertouzos proposed this concept with banking as the application background.
Compared with general encryption schemes that focus on data storage security, the most interesting thing about homomorphic encryption schemes is that they focus on data processing security.
Specifically, homomorphic encryption provides a function for encrypting private data. In the homomorphic encryption scheme, other participants can process the private data, but the processing will not leak any original content. At the same time, the user who has the key can decrypt the processed data and the result is exactly the correct data after processing.
For example, ALICE bought a piece of gold and wanted a worker to make it into a necklace. Is there a way that allows the worker to process the gold but not get any gold?
To solve this problem, ALICE can lock the gold bars in a sealed box with a single key. The box has two holes, with a glove installed in each hole. Workers can handle the gold bars inside the box while wearing gloves without being able to steal any gold bars.
After the processing was completed, ALICE took the entire box back, opened the lock, and got the processed necklace.
Here, the box corresponds to the fully homomorphic encryption algorithm, and the worker processing corresponds to the execution of homomorphic operations. When the data cannot be obtained, the encryption result is directly processed.
Fully Homomorphic Encryption Application Scenarios
In Web2, homomorphic encryption is almost tailor-made for cloud computing. Consider the following scenario: a user wants to process a piece of data, but his computer has weak computing power and cannot get the result in time. Then the user can use the concept of cloud computing to let the cloud help him process the data and get the result.
However, if the data is directly handed over to the cloud, security cannot be guaranteed. So he can first encrypt the data using homomorphic encryption, and then let the cloud directly process the encrypted data and return the processing results to him.
In this way, the user pays the cloud service provider, gets the processing results, and the cloud service provider earns the fee. However, fully homomorphic encryption also has the disadvantage of being limited by computing power:
High computational cost: Fully homomorphic encryption requires more complex mathematical algorithms and larger ciphertexts than traditional encryption, which makes it slower and more resource-intensive to perform operations on encrypted data.
Low computational efficiency: FHE (Fully Homomorphic Encryption) only supports arithmetic operations on encrypted data, such as addition, multiplication, and exponential operations. For more complex functions such as sorting, searching, or string operations, more tedious processing is required before execution. High computing power requirements.
Fortunately, we are in an era of explosive computing power. With the advancement of FHE and Web3 development, computing power performance and cost are expected to match the requirements of FHE. Therefore, this is a good time to ambush the FHE track.
Fhenix
Fhenix is the first blockchain to use fully homomorphic encryption technology, which can provide encrypted data computing capabilities for EVM smart contracts.
The fhEVM used by Fhenix was originally developed by Zama, a cryptography company that builds open source encryption solutions for blockchain and artificial intelligence, and was integrated with Fhenix Network after a strategic partnership.
In addition, Fhenix also uses Arbitrum's Nitro validator and Zama's fully homomorphic ring encryption rust library tfhe-rsr. This shows the close relationship between Zama and Fhenix.
Zama's official website shows that the company is providing FHE-based Web3 solutions for some cutting-edge Web2 use cases, such as face recognition, voice recognition, and smart contracts (which is what Fhenix is currently doing). In the future, we can expect Zame to integrate all these applications into the Fhenix ecosystem.
In September this year, Fhenix raised $7 million in a seed round of financing, led by Multicoin Capital and Collider Ventures, with participation from Node Capital, Bankless, HackVC, TaneLabs, Metaplanet, and Tarun Chitra and Robert Leshner's Robot Ventures.
Compared to zk, which can only verify the data segments encrypted by it, cannot merge private data from multiple parties, and therefore cannot facilitate most cryptographic computations, FHE allows for a higher level of data security and supports unprecedented use cases through its "holistic" encryption capabilities.
Therefore, the ability to have privacy on Fhenix will not only solve privacy issues, but also pave the way for hundreds of new use cases - blind auctions, on-chain identity verification and KYC, tokenization of real-world assets, private voting for DAOs, etc.
04 Summary: Comparison between ZK and FHE
After learning about ZK and FHE, two cutting-edge privacy smart contract solutions, many readers are still confused about the two technical routes of zero-knowledge proof (ZK) and fully homomorphic encryption.
The differences between the two, in addition to the encryption flexibility mentioned above, are also reflected in:
To summarize from a technical perspective, ZK focuses on proving correctness while protecting the privacy of statements; FHE focuses on performing calculations without decryption, protecting the privacy of data.
From the perspective of blockchain industry development, projects using ZK technology were developed earlier, from ZCash, which only has transfer functions, to the zkVM blockchain that supports smart contracts and is currently under development. Compared with FHE, there is more blockchain industry technology accumulation; and the FHE theory was born much later than ZK. It is a hot topic in academia. Only recently have Web3 projects using FHE technology for financing appeared, so its development started slower than ZK.
The common point between the two is that they both rely on the development of computing power, and the development of the privacy track has benefited from the explosion of computing power. It is also thanks to the improvement of computing power in recent years that these cutting-edge technologies can truly be exposed to users.
references
[01] Beyond ZK: The Definitive Guide to Web3 Privacy (Part 2) https://scrt.network/blog/beyond-zk-guide-to-web3-privacy-part-2/
[02] Introduction to FHE: What is FHE, how does FHE work, how is it connected to ZK and MPC, what are the FHE use cases in and outside of the blockchain, etc. https://taiko.mirror.xyz/2O9rJeB-1PalQeYQlZkn4vgRNr_PgzaO8TWUOM5wf3M
[03] Ola: A ZKVM-based, High-performance and Privacy-focused Layer2 platform https://ethresear.ch/t/ola-a-zkvm-based-high-performance-and-privacy-focusedlayer2-platform/15248
[04] FHE-Rollups: Scaling Confidential Smart Contracts On Ethereum And Beyond – Whitepaper https://www.fhenix.io/fhe-rollups-scaling-confidential-smart-contracts-on-ethereum-and-beyond-whitepaper/
