• Dhavan.S.Gowda
• Maithreyan S M
• Paplesh
• Pavan P
• Dr. Srinivasa A H

The rapid evolution of Artificial Intelligence (AI) and Natural Language Process- ing (NLP) has enabled transformative capabilities in legal technology. This paper presents an AI-Powered Virtual Legal Assistant that automates (i) legal PDF analysis and summariza- tion, (ii) retrieval-augmented, citation-backed answering using live Google Search and domain allow-listing, and (iii) verdict prediction through reasoning over case summaries and precedent patterns. The system integrates a Next.js front-end, Node/TypeScript APIs, pdf-parse for extrac- tion, a local Ollama–Mistral LLM for summarization and reasoning, and Hugging Face utilities for embeddings. Evaluated across 50 Indian matters, the assistant achieved 96% extraction suc- cess for text-PDFs, average summary accuracy of 9/10, and overall verdict-direction accuracy of ∼72%. The solution demonstrates a practical, privacy-conscious pathway for enhancing legal research, pedagogy, and decision support in the Indian judicial context.

Legal AI; NLP; Retrieval-Augmented Generation; Legal Document Summariza- tion; Verdict Prediction; Indian Judiciary; LegalTech; Large Language Models.

"AI-powered Virtual Legal Assistant for Case Analysis and Verdict Prediction", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 11, page no.f407-f412, November-2025, Available :http://www.jetir.org/papers/JETIR2511569.pdf

"AI-powered Virtual Legal Assistant for Case Analysis and Verdict Prediction", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 11, page no. ppf407-f412, November-2025, Available at : http://www.jetir.org/papers/JETIR2511569.pdf