Grass is designed to separate user identity and personal activity from the bandwidth-sharing process. When someone installs the Grass application or extension, the traffic being routed through their connection is limited to publicly accessible data. This includes static webpages, public APIs, and open-source media content. Requests are filtered to ensure they don’t include user-specific sessions, accounts, or information tied to personal browsing behavior.
The network architecture is structured so that personal data on the contributor’s device is not part of the operation. Grass does not request access to browsing history, local files, or private cookies. Its routing logic avoids login-based platforms and authenticated endpoints, instead focusing on collecting information that is openly available on the public internet. These design choices reflect a strict scope: the protocol exists to support AI training data, not to observe user behavior.
Participants are given clear documentation explaining how data flows through the system and how privacy is maintained. Users also have access to built-in settings that allow them to disable or pause their participation at any time. This opt-in structure is reinforced by transparency in how public data is selected, categorized, and handled. The goal is to allow people to contribute with confidence, knowing the technical limits placed on what the system can access.
Encryption is used throughout the Grass network to protect communications, maintain data integrity, and reduce exposure to interference. Whether traffic is flowing between Nodes and Routers or from Routers to Validators, it is encrypted in transit. This ensures that external actors cannot easily intercept or alter the information being transferred. The same applies to system logs and internal communication across the network infrastructure.
To strengthen reliability, Grass commissions external security audits from established firms that specialize in Web3 and distributed systems. These reviews analyze smart contracts, system architecture, and application layers for vulnerabilities. Any identified weaknesses are documented and addressed before deployment. Public summaries of the audits are made available to provide visibility into how the network manages security risks.
Grass also collaborates with independent researchers and infrastructure experts to review key updates and critical systems on an ongoing basis. This approach allows the protocol to stay aligned with current best practices in cybersecurity. The team emphasizes preemptive action by keeping software dependencies updated, patching known vulnerabilities, and applying layered security models.
Grass gives users full control over when and how they participate in the network. When the application or browser extension is installed, participation is off by default until the user activates it. Once running, users can limit how much bandwidth they contribute, restrict usage to specific times of day, or shut down participation entirely with one click. This control structure applies across both desktop and browser versions, with additional advanced settings available on the desktop app.
Bandwidth settings can be adjusted at any time based on user preference or device conditions. Grass includes real-time dashboards that show contribution statistics, estimated rewards, and the types of requests handled. This gives users the ability to make informed decisions and adapt their participation based on personal use or system load. There is no requirement to keep the app active, and rewards are calculated only for completed and verified traffic.
Grass also provides transparency around the kind of content being requested. Since all requests are limited to public domains, users can contribute without concern about being linked to private services, tracked sessions, or authenticated platforms. The combination of access control, traffic transparency, and optional participation makes the experience predictable and adaptable. Users contribute only when and how they decide, with full visibility into their role in the system.
Highlights
Grass is designed to separate user identity and personal activity from the bandwidth-sharing process. When someone installs the Grass application or extension, the traffic being routed through their connection is limited to publicly accessible data. This includes static webpages, public APIs, and open-source media content. Requests are filtered to ensure they don’t include user-specific sessions, accounts, or information tied to personal browsing behavior.
The network architecture is structured so that personal data on the contributor’s device is not part of the operation. Grass does not request access to browsing history, local files, or private cookies. Its routing logic avoids login-based platforms and authenticated endpoints, instead focusing on collecting information that is openly available on the public internet. These design choices reflect a strict scope: the protocol exists to support AI training data, not to observe user behavior.
Participants are given clear documentation explaining how data flows through the system and how privacy is maintained. Users also have access to built-in settings that allow them to disable or pause their participation at any time. This opt-in structure is reinforced by transparency in how public data is selected, categorized, and handled. The goal is to allow people to contribute with confidence, knowing the technical limits placed on what the system can access.
Encryption is used throughout the Grass network to protect communications, maintain data integrity, and reduce exposure to interference. Whether traffic is flowing between Nodes and Routers or from Routers to Validators, it is encrypted in transit. This ensures that external actors cannot easily intercept or alter the information being transferred. The same applies to system logs and internal communication across the network infrastructure.
To strengthen reliability, Grass commissions external security audits from established firms that specialize in Web3 and distributed systems. These reviews analyze smart contracts, system architecture, and application layers for vulnerabilities. Any identified weaknesses are documented and addressed before deployment. Public summaries of the audits are made available to provide visibility into how the network manages security risks.
Grass also collaborates with independent researchers and infrastructure experts to review key updates and critical systems on an ongoing basis. This approach allows the protocol to stay aligned with current best practices in cybersecurity. The team emphasizes preemptive action by keeping software dependencies updated, patching known vulnerabilities, and applying layered security models.
Grass gives users full control over when and how they participate in the network. When the application or browser extension is installed, participation is off by default until the user activates it. Once running, users can limit how much bandwidth they contribute, restrict usage to specific times of day, or shut down participation entirely with one click. This control structure applies across both desktop and browser versions, with additional advanced settings available on the desktop app.
Bandwidth settings can be adjusted at any time based on user preference or device conditions. Grass includes real-time dashboards that show contribution statistics, estimated rewards, and the types of requests handled. This gives users the ability to make informed decisions and adapt their participation based on personal use or system load. There is no requirement to keep the app active, and rewards are calculated only for completed and verified traffic.
Grass also provides transparency around the kind of content being requested. Since all requests are limited to public domains, users can contribute without concern about being linked to private services, tracked sessions, or authenticated platforms. The combination of access control, traffic transparency, and optional participation makes the experience predictable and adaptable. Users contribute only when and how they decide, with full visibility into their role in the system.
Highlights