TraceMonitor: A Novel AI-Based Chrome Extension to Enhance User Awareness of Privacy Leakage towards Green Web Browsing

With hundreds of millions of people utilizing the Internet each day, companies have massive incentives to collect and sell the personal data of users navigating their webpages and platforms. Many companies track, collect, and record user data to be used in targeted advertisements and sold to third-party companies, oftentimes without informing the user explicitly. Because of this, it is difficult for users to detect whenever their personal data is being tracked or collected. Towards creating a green web browsing environment in terms of privacy, this paper presents a novel Chrome web extension designed to inform users of: 1) which companies are potentially tracking their data; and 2) which specific personal identifiers (i.e., IP address, name, zip code, etc.) these companies could be collecting. It enables users to track their privacy footprints in online browsing (similar to carbon footprint of energy usage). Specifically, the tool identifies private information trackers by analyzing the source code of a website and third-party companies' analytics scripts. After downloading the entire source code of a website, OpenAI GPT 3.5 is utilized to identify company names present in links and external files that are being used by the website. Then, personal identifiers possibly being collected are identified through checking the data analytics Javascript files present on a website. Tested on 40 websites spanning across five website categories, including food delivery, entertainment, travel, government, and news websites, the extension showed an average prediction accuracy of 91.03%.