PrivaSee AI: A Generative AI Framework for Dynamic Web Tracking Detection and Explainable Privacy Auditing

Modern websites use third-party scripts, cookies, and other tracking mechanisms that may collect information about users' browsing activities. Understanding these activities through conventional browser developer tools can be difficult for non-technical users. This paper presents PrivaSee AI, a prototype framework for dynamic web privacy auditing and explainable privacy analysis. The system uses Puppeteer and headless Chromium to observe website behaviour, collect cookies, enumerate scripts, and inspect selected HTTP security headers, including Content-Security-Policy, Strict-Transport-Security, and X-Frame-Options. The collected telemetry is processed using Google Gemini 2.5 Flash to generate structured privacy explanations, including a privacy grade, privacy story, creepiness report, and security assessment. The authors report an exploratory evaluation across 50 websites from five industry sectors and a usability study involving 15 university students. The reported results indicate an improvement in mean self-rated comprehension from 1.87 to 4.73 on a five-point scale. The findings are preliminary and require verification against the original experimental records. PrivaSee AI is intended to improve privacy awareness and explainability rather than provide a complete security or legal compliance assessment. Keywords: Explainable AI, Web Privacy, Dynamic Tracking Detection, Puppeteer, Chromium, Gemini, Browser Telemetry, Privacy Auditing.

Authors

Institutions

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-15
DOI
https://doi.org/10.5281/zenodo.24241127
Primary Topic
Privacy, Security, and Data Protection
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

PrivaSee AI: A Generative AI Framework for Dynamic Web Tracking Detection and Explainable Privacy Auditing

M.Geetha, L. Jaya Sai Reddy
Zenodo (CERN European Organization for Nuclear Research)
Privacy, Security, and Data Protection
article

PrivaSee AI: A Generative AI Framework for Dynamic Web Tracking Detection and Explainable Privacy Auditing

M.Geetha, L. Jaya Sai Reddy
article en

Abstract

Modern websites use third-party scripts, cookies, and other tracking mechanisms that may collect information about users' browsing activities. Understanding these activities through conventional browser developer tools can be difficult for non-technical users. This paper presents PrivaSee AI, a prototype framework for dynamic web privacy auditing and explainable privacy analysis. The system uses Puppeteer and headless Chromium to observe website behaviour, collect cookies, enumerate scripts, and inspect selected HTTP security headers, including Content-Security-Policy, Strict-Transport-Security, and X-Frame-Options. The collected telemetry is processed using Google Gemini 2.5 Flash to generate structured privacy explanations, including a privacy grade, privacy story, creepiness report, and security assessment. The authors report an exploratory evaluation across 50 websites from five industry sectors and a usability study involving 15 university students. The reported results indicate an improvement in mean self-rated comprehension from 1.87 to 4.73 on a five-point scale. The findings are preliminary and require verification against the original experimental records. PrivaSee AI is intended to improve privacy awareness and explainability rather than provide a complete security or legal compliance assessment. Keywords: Explainable AI, Web Privacy, Dynamic Tracking Detection, Puppeteer, Chromium, Gemini, Browser Telemetry, Privacy Auditing.

Zenodo (CERN European Organization for Nuclear Research)
Rajiv Gandhi University of Knowledge Technologies (IN)
Privacy, Security, and Data Protection
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon