Grass Vocabulary: Key Terms Explained
Want to dive deeper into Grass and understand its operations more comprehensively? The sections below will break down key terms, explaining what they mean, why they matter, and how they connect to Grass.
| Term | What it means | Why you should care |
|---|---|---|
| Internet Bandwidth | The speed and capacity of your internet connection; how much data can be sent or received at once. | Your network connection isn’t always fully utilized. Picture a multi-lane highway. Most of the time, some lanes are empty. That unused capacity is your unused internet bandwidth. Grass creates an ethical proxy network, and when you share your unused internet bandwidth with verified institutions, you’re rewarded, enabling them to access the web for AI training or to enhance their online products and services with public web data. |
| Proxy | A device connected to the internet that sends web requests on behalf of others. | Businesses, marketers, and researchers use proxies to route their web requests through real residential IP addresses, making their activity appear as if it’s coming from a regular internet user at home. This helps them access geo-restricted content, verify ads, scrape web data, test localized online experiences, and protect privacy. |
| Residential Proxy Network | A network of people’s internet connections used to route web requests for others. | Many free apps, VPNs, and background services secretly turn people’s home internet connections into residential proxies without clear consent or payment. Those companies then act as middlemen, selling the proxies to businesses, marketers, and researchers so they can use your internet bandwidth to access the web. This means your internet connection is used by others without your knowledge, slowing your speeds and exposing you to security risks. |
| Ethical Proxy | An internet connection shared with full user consent and fair compensation. | Unlike companies that secretly sell your internet bandwidth through free apps and VPNs with hidden terms, Grass lets users voluntarily share unused internet bandwidth and get fairly compensated. This ensures Grass proxies are ethically sourced, fostering a fair and transparent digital ecosystem. When your internet connection becomes a Grass node, it means you’ve allowed your internet connection to be used as an ethical proxy. |
| Personal Data | Information about you or what you do on the internet. | When you connect to Grass, you only share unused internet bandwidth, never any personal information or your online activities. Sharing unused internet bandwidth is like letting a friend use your home’s WiFi with their phone. The only difference is that with Grass, you get rewarded for sharing bandwidth. |
| Public Web Data | Content on the internet that you can view without logging in. | Public web data powers search engines, market research, price comparisons, and AI training. |
| Value Extraction | Taking value from someone without giving fair return. | Many online services make money off your activity, data, or attention without sharing any value with you. Social media platforms, search engines, and ad networks extract billions of dollars from user-generated content and behavior, while users get little in return. Understanding value extraction helps you see who really benefits from the services you use and whether there are fairer alternatives. |
| Web Request | A message your device sends to a website to ask it to load. | When you browse the internet, your device constantly sends web requests to websites to load the text, images, and videos on a page so you can access them. |
| IP Address | An IP address is like your device’s street address on the internet. It helps websites, apps, and networks know who is making a web request so they know where to send data. | Your IP address determines your online identity, location, and access to online services. It impacts privacy, security, and connectivity and can be used to track, block, or grant access to content. It’s why a New Yorker sees U.S. Netflix shows while a Parisian sees French Netflix shows. |
| Node | An individual device in a residential IP proxy network. | Each node forms part of the Grass infrastructure, enabling the network to provide proxy services. When you sign up for Grass, your device becomes a “node,” sharing unused internet bandwidth with the network. |
| Web Scraping | Finding and collecting information stored on the public web. | Every time you search for flight prices, compare products, or check reviews, it’s likely that companies have used web scraping to collect that information. While big companies scrape data to improve services, some block scraping to control valuable information. This creates a power imbalance, where only a few companies benefit from public web data instead of it being accessible to everyone. |
| Dataset | A collection of information that can be used to find patterns, make predictions, or generate insights. It can be as simple as a list of soccer stats or as complex as a massive collection of medical images used to train AI. | Understanding datasets helps people recognize how decisions in AI, business, and research are shaped by the data used. Datasets power everything from search engines to AI assistants and financial models. The quality of a dataset impacts the accuracy of AI models, meaning biased or incomplete data can lead to flawed predictions or unfair outcomes. |
| Multimodal Data | A mix of different types of data, like text, images, and sound. | Multimodal data enables richer insights, enhancing decision-making in fields like AI, healthcare, and autonomous systems. It enables more robust natural language processing, more accurate medical diagnoses, and better human-machine interactions, improving the adaptability and intelligence of technology. |
| AI Training | Feeding massive amounts of data to an AI model so it can learn relationships and get better at responding to your prompts. | The quality of an AI depends on the data it’s trained on. If the training data is biased, outdated, or controlled by a few companies, the AI model will reflect those same issues. AI training impacts everything from chatbots to voice assistants and even self-driving cars. |