AI-Powered Contract Compliance Checker

AI-Powered Contract Compliance Checker

Rahul R N, Monish V, Akshaya R S

Rahul R N – B.tech AI/ML – Sri Shakthi Institute of Engineering & Technology

Monish V – B.tech AI/ML – Sri Shakthi Institute of Engineering & Technology

Akshaya R S – B.tech AI/ML – Sri Shakthi Institute of Engineering & Technology

Abstract - This project presents an AI-powered system designed to automate contract compliance checking using advanced natural language processing and machine learning techniques. The primary objective is to ensure that legal agreements adhere to specified compliance requirements, reducing manual review time and minimizing the risk of oversight. The system employs a retrieval-augmented generation (RAG) framework, where contract data is converted into vector embeddings using ChromaDB and then queried for relevant context. A Streamlit-based user interface allows users to upload contracts and view compliance insights. Text extraction and clause identification are performed through a FastAPI backend integrated with a large language model via the Groq API. The system effectively identifies missing or non-compliant clauses by comparing input contracts against a preprocessed dataset. Results demonstrate the tool’s ability to detect inconsistencies with high accuracy, offering significant potential for legal and corporate environments. This solution streamlines the compliance review process, increases efficiency, and provides real-time AI-driven suggestions to improve contractual documents.

Key Words: Contract compliance, Natural Language Processing, ChromaDB, Retrieval-Augmented Generation, Legal AI, Streamlit.