Grass Network: Reshaping the Internet Incentive Mechanism

Grass is a decentralized network scraping platform deployed on the Solana chain, combining AI, Depin, and Solana technology. Its goal is to become the data layer for AI, helping businesses and organizations train AI models by utilizing users' idle internet bandwidth.

The core concept of Grass is to redefine the incentive structure of the internet, allowing users to benefit directly from the network. Users can earn Grass Points as rewards by sharing unused bandwidth resources. Currently, the network has over 2 million user-operated nodes that have captured a large amount of data for AI models.

Technical Architecture

Grass has built a sovereign data Rollup network on Solana, enabling it to handle the entire process from data collection to validation and dataset construction. The network mainly consists of the following parts:

1. Validator (: Receives, verifies, and batch processes web transactions from the router, generating ZK proofs to verify on-chain session data.

2. Router ): Connects Grass nodes and validators, maintains network traceability, and relays bandwidth.

3. Grass Node (Node ): Utilizing users' idle bandwidth to fetch public web data, node operators are rewarded based on the amount of relay data.

4. ZK Processor: Validity proof of batch processing Web request session data, and submission to the blockchain.

5. Grass Data Ledger: Stores the fetched data, links the data to the chain for proof, ensuring data traceability.

6. Edge embedding model: Converts unstructured web data into structured data usable by AI models.

Technical Features

The Grass network is located between the client and the web server. The client's web requests are sent through validators and ultimately routed by Grass nodes. The server responds to the requests, allowing data to be fetched and returned. After being cleaned and processed, the data is used to train AI models.

Grass data ledger is a permanent storage location for data, embedding metadata to record data lineage. The metadata proof for each dataset is stored on Solana.

The ZK processor helps record the source of data sets. When nodes send web requests, the returned encrypted response contains the origin information of the data set. This information is crucial for the correct training of AI models. The ZK processor also protects the privacy of on-chain settlement data and improves transaction processing capacity through batch processing.

Metadata not only records the source of the data, but also indicates routing nodes, allowing nodes to receive rewards based on contributions while protecting identity privacy. This mechanism encourages more users to join the network, expanding data scraping capabilities.

Node Operation and Security Mechanism

Running a Grass node is free, and node operators receive rewards based on the relay traffic. Nodes are mainly used to pass Web requests and return encrypted server responses.

Node reputation scoring includes aspects such as integrity, consistency, timeliness, and availability.

In terms of security, Grass only routes internet traffic and does not access user personal data. The network uses bandwidth encryption to protect users and collaborates with AppEsteem for 24-hour security monitoring.

Grass Token Functions

Grass token holders can participate in network transactions, purchase datasets, stake to earn rewards, and engage in network governance. Currently, the annualized staking yield is approximately 45%, with about 33% of the tokens participating in staking.

Router Staking and Earnings

The router acts as a decentralized hub connecting network nodes and managing requests. Users can stake Grass to the Router to earn rewards, and the commission rates of different Routers vary.

Summary

Grass is committed to building a fair and open decentralized data layer, addressing the ethical and quality issues of current internet data extraction. Its unique data Rollup architecture and metadata mechanism enhance data transparency and create a fair incentive mechanism for node providers. As an important player in the intersection of Web3 and AI, Grass provides an open data infrastructure for AI companies and protocols, with broad development prospects.