POLICY GUARDIAN: A HYBRID TWO-TIER AI FRAMEWORK FOR AUTOMATED LEGAL DOCUMENT DISCOVERY, RISK CLASSIFICATION, AND SEMANTIC SIMPLIFICATION

May 2026
Vol-12, Issue-3
Paper ID: 28422
ISSN: 2395-4396
Downloads: 141
Download PDF

Abstract & Details

Research Area
Computer Science & Engineering
Keywords
Legal NLP Terms of Service Unfair Clause Detection DistilBERT Large Language Models Chrome Extension Consumer Protection Privacy by Design Real-Time Analysis Transformer Models
Abstract
Every day, billions of internet users click "I Agree" on Terms of Service and Privacy Policy agreements without reading a single word. This is not a matter of carelessness — it is a structural impossibility imposed by documents deliberately written in dense legal language that is practically inaccessible to ordinary people. The consequences are significant: users routinely and unknowingly waive their rights to legal recourse, consent to unlimited data sharing, and accept liability terms they would reject if they understood them. Policy Guardian is a real-time, AI-powered browser extension engineered to close this transparency gap. The system deploys a novel Hybrid Two-Tier AI architecture: a locally fine-tuned DistilBERT transformer model performs high-speed, privacy-preserving classification of individual legal clauses directly on the user's machine, while a cloud-based large language model (Llama 3.3 70B via the Groq API) generates consolidated, plain-English verdicts for the flagged risky clauses only. The system also implements a Non-Destructive DOM Cloning scraper and a semantic Auto-Finder algorithm to automatically locate and extract legal documents from any website without user intervention. The fine-tuned Policy-Guardian-BERT model was trained on a rigorously cleaned subset of the lex_glue/unfair_tos benchmark dataset, achieving a classification accuracy of 96.33% and a macro-averaged F1-score of 0.9545 on a held-out validation set. The complete end-to-end pipeline — from link discovery and text extraction through risk classification to plain-English verdict generation — completes in approximately 1.75 seconds, making it entirely practical for everyday browsing. The system was deployed as a Chrome Extension using the Manifest V3 specification, connected to a Flask REST API backend, and validated across real-world websites including Reddit, Pinterest, and a purpose-built demonstration platform. By performing primary classification locally, ninety percent of the user's legal text is analyzed with zero data transmission to external servers, realizing a genuine Privacy by Design architecture. This paper presents the complete system design, training methodology, evaluation results, and a discussion of the social and ethical implications of deploying AI as a real-time consumer protection tool in the legal domain.

Author Information

# Name Institute / Affiliation
1 Ashik James Holy Grace Academy of Engineering
2 Dhrupadh VS Holy Grace Academy of Engineering
3 Jeswin Joy Holy Grace Academy of Engineering
4 Jishnu Manoj Holy Grace Academy of Engineering
5 Sony K T Holy Grace Academy of Engineering
6 Sanam E Anto Holy Grace Academy of Engineering

How to Cite

Use the following formats to cite this article in your research.

APA Style
James, Ashik, VS, Dhrupadh, Joy, Jeswin, Manoj, Jishnu, T, Sony K, & Anto, Sanam E (2026). POLICY GUARDIAN: A HYBRID TWO-TIER AI FRAMEWORK FOR AUTOMATED LEGAL DOCUMENT DISCOVERY, RISK CLASSIFICATION, AND SEMANTIC SIMPLIFICATION. International Journal of Advance Research and Innovative Ideas In Education, 12(3), 314-322.
MLA Style
James, Ashik, et al. "POLICY GUARDIAN: A HYBRID TWO-TIER AI FRAMEWORK FOR AUTOMATED LEGAL DOCUMENT DISCOVERY, RISK CLASSIFICATION, AND SEMANTIC SIMPLIFICATION." International Journal of Advance Research and Innovative Ideas In Education, vol. 12, no. 3, 2026, pp. 314-322.
IEEE Style
Ashik James, Dhrupadh VS, Jeswin Joy, Jishnu Manoj, Sony K T, and Sanam E Anto, "POLICY GUARDIAN: A HYBRID TWO-TIER AI FRAMEWORK FOR AUTOMATED LEGAL DOCUMENT DISCOVERY, RISK CLASSIFICATION, AND SEMANTIC SIMPLIFICATION," International Journal of Advance Research and Innovative Ideas In Education, vol. 12, no. 3, pp. 314-322, 2026.
Vancouver Style
James Ashik, VS Dhrupadh, Joy Jeswin, Manoj Jishnu, T Sony K, Anto Sanam E. POLICY GUARDIAN: A HYBRID TWO-TIER AI FRAMEWORK FOR AUTOMATED LEGAL DOCUMENT DISCOVERY, RISK CLASSIFICATION, AND SEMANTIC SIMPLIFICATION. International Journal of Advance Research and Innovative Ideas In Education. 2026;12(3):314-322.
Harvard Style
James, Ashik, VS, Dhrupadh, Joy, Jeswin, Manoj, Jishnu, T, Sony K, & Anto, Sanam E (2026) 'POLICY GUARDIAN: A HYBRID TWO-TIER AI FRAMEWORK FOR AUTOMATED LEGAL DOCUMENT DISCOVERY, RISK CLASSIFICATION, AND SEMANTIC SIMPLIFICATION', International Journal of Advance Research and Innovative Ideas In Education, 12(3), pp. 314-322.
Chicago Style
James, Ashik, et al. "POLICY GUARDIAN: A HYBRID TWO-TIER AI FRAMEWORK FOR AUTOMATED LEGAL DOCUMENT DISCOVERY, RISK CLASSIFICATION, AND SEMANTIC SIMPLIFICATION." International Journal of Advance Research and Innovative Ideas In Education 12, no. 3 (2026): 314-322.
Turabian Style
James, Ashik, et al. "POLICY GUARDIAN: A HYBRID TWO-TIER AI FRAMEWORK FOR AUTOMATED LEGAL DOCUMENT DISCOVERY, RISK CLASSIFICATION, AND SEMANTIC SIMPLIFICATION." International Journal of Advance Research and Innovative Ideas In Education 12, no. 3 (2026): 314-322.

Export Citation

Related Research

PReFace : AI -Assisted Plausible Facial Restoration and Damage Reconstruction System
Syeda Arbeena Kausar et al. 2026 Computer Science and Engineering (Computer Vision, Deep Learning and Image Processing)
Download PDF
Customer Churn Prediction Using Machine Learning
Dr. Roshni Khlhe et al. 2026 Machine Learning, Artificial Intelligence, Data Analytics, and Customer Churn Prediction.
Download PDF
Automated web vulnerability scanner
Jitendra Sahu et al. 2026 Computer science
Download PDF
BLOOD BANK DEMAND PREDICTION SYSTEM: MACHINE LEARNINGBASED BLOOD DEMAND FORECASTING
Uday Dharmapal Wankhede et al. 2026 Computer Science & Engineering
Download PDF