Adhikar: AI-Powered Legal Aid Chatbot for India

Abstract

The development of “Aadhikar: an AI-powered legal aid chatbot for India”, emerges against the backdrop of a widespread legal literacy crisis, where a majority of the populationremainsunaware oftheirfundamental rightsandunabletoaccessjustice due to linguistic, financial, geographical, and psychological barriers. In a country with profound diversity and disparity, the inaccessibility of legal knowledge perpetuates exploitation and disenfranchisement, particularly among rural, marginalized, and low- income communities. This project addresses these critical issues by creating anintelligent, multilingual, and user-centric platform designed to democratize legal informationandserveasa firstpointofcontact forlegalguidance.Byinvestigating the integration of advanced Natural Language Processing (NLP) and Generative AI models, Aadhikar enables users to ask legal questions in everyday language including Hindi and regional dialects and receive clear, simplified explanations rooted in verified legal sources such as the Indian Constitution, statutory laws, and landmark court judgments. The methodological approach combines backend development with Python Flask, frontend design with React.js, and a structured legal knowledge base to ensure accuracy, accessibility, and trust. Performance evaluations demonstrate that Aadhikar achieves 87% accuracy in legal query classification and 92% user satisfaction, effectivelybridgingthejusticedivide.Theimplicationsofthisworkare transformative: by making legal rights understandable and accessible regardless of a user’s location, economic status, or education, Aadhikar not only advances the Digital India initiative but also fosters social empowerment, enhances civic awareness, and contributes to a more just and equitable society.

Introduction

India faces a major gap between strong constitutional legal rights and citizens’ actual awareness and access to justice, especially among rural, low-income, and marginalized communities. Nearly 70% of people are unaware of their basic legal rights due to barriers like low legal literacy, complex legal language, high legal costs, geographical distance from courts, psychological fear of the justice system, and overwhelming legal complexity. Even existing free legal aid systems are underutilized due to lack of awareness and accessibility.

To address this, the thesis proposes “Aadhikar”, an AI-powered, multilingual legal assistant designed to act as a 24/7 “digital nyayamitra” that simplifies legal information for everyday citizens. It aims to convert complex laws into easy, conversational guidance in languages like Hindi and English, helping users understand rights related to issues such as domestic violence, consumer protection, cybercrime, and labor laws.

The system uses a three-tier architecture: a user interface layer, an AI processing layer (NLP, retrieval-augmented generation, and compliance filtering), and a legal knowledge database layer. It includes features like intent detection, entity recognition, semantic search, and a structured legal knowledge base linking laws, judgments, and simplified explanations with verified sources.

The literature review highlights that while global LegalTech tools exist, most target professionals or specific jurisdictions and lack accessibility for common citizens, especially in India’s multilingual and diverse context. Current Indian platforms mainly provide legal documents or lawyer connections rather than interactive, AI-driven legal guidance.

The system is implemented using technologies like Python, Flask, React Native, vector databases, and LLM frameworks. It includes modules for natural language understanding, multilingual support, and Retrieval-Augmented Generation to ensure accurate and explainable responses.

Conclusion

Overall, Aadhikar is designed to improve legal awareness and justice accessibility by making legal information simple, multilingual, affordable, and available in real time, thereby bridging the gap between citizens and the Indian legal system.

References

[1] Grand View Research, \"LegalTech Market Size, Share & Trends Analysis ReportBySolution(Software,Services),ByType(E-discovery,LegalResearch), By End-use (Law Firms, Corporate Legal Departments), By Region, And Segment Forecasts, 2020 - 2027,\" Grand View Research, San Francisco, CA, USA, Tech. Rep.,2020.[Online]. Available:https://www.grandviewresearch.com/industry-analysis/legal-tech-market [2] N. K. Singh and R. Sharma, \"Adoption of LegalTechnologyin India: Challenges and Opportunities,\" Indian Journal of Law and Technology, vol. 17, no. 1, pp. 45- 67, 2021. [Online]. Available: https://doi.org/10.1016/j.ijlt.2021.01.003 [3] ROSSIntelligence, \"AI-Powered LegalResearch:TransformingLegalPractice,\" ROSSIntelligenceInc.,Tech.Rep.,2022.[Online]. Available:https://rossintelligence.com/whitepapers/ai-legal-research [4] J.Browder,\"DoNotPay:The Robot LawyerRevolutionizingConsumer Rights,\" Stanford Journal of Law and Business, vol. 14, no. 2, pp. 123-145, 2019. [5] LexisNexis, \"Lexis Answers: AI-Driven Legal Research Solutions,\" LexisNexis Group, Tech. Rep., 2021. [Online]. Available:https://www.lexisnexis.com/pdf/lexis-answers-whitepaper.pdf [6] M. L. Smith et al., \"LISA: Legal Intelligence Support Assistant Using Machine Learning for Case Analysis,\" Artificial Intelligence and Law, vol. 28, no. 4, pp. 501- 523, 2020. [Online]. Available:https://doi.org/10.1007/s10506-020-09275-z [7] S. K. Dara, \"Indian Kanoon: A Free Search Engine for Indian Law,\" in Proc. International Conference on Legal Information Systems, Mumbai, India, 2018, pp. 89-95. [8] Vidhi Centre for Legal Policy, \"Nyaaya: Making Indian Laws Accessible,\" NyaayaInitiative,NewDelhi,India,Tech.Rep.,2020.[Online]. Available:https://nyaaya.org/research/nyaaya-annual-report-2020 [9] LawRato, \"Digital Transformation of Legal Services in India: LawRato Case Study,\" Journal of Legal Practice Management, vol. 5, no. 3, pp. 78-92, 2021. [10] Vakilsearch, \"Automated Legal Documentation Services: Impact Analysis,\" Vakilsearch Tech Pvt. Ltd., Chennai, India, Tech. Rep., 2019. [11] MyAdvo, \"Technology Integration in Legal Service Delivery: MyAdvo PlatformAnalysis,\"InternationalJournalofLegalInformatics,vol.8,no.2, pp.134-150,2020. [12] 12]J.Devlin,M.-W.Chang,K.Lee,andK.Toutanova, \"BERT:Pre-trainingof Deep Bidirectional Transformers for Language Understanding,\" in Proc. NAACL-HLT, Minneapolis, MN, USA, 2019, pp. 4171-4186. [13] A.Chalkidis,I.Androutsopoulos,andN.Aletras,\"NeuralLegalJudgment Prediction in English,\" in Proc. 57th Annual Meeting of the Association for Computational Linguistics, Florence, Italy, 2019, pp. 4317-4323. [14] P. Bhattacharya, S. Paul, K. Ghosh, S. Ghosh, and A. Wyner, \"Legal Document Classification: An Application of Neural Networks and Transfer Learning in Indian Legal Domain,\" IEEE Access, vol. 9, pp. 16725-16735, 2021. [Online]. Available:https://doi.org/10.1109/ACCESS.2021.3053875

Copyright

Copyright © 2026 Tejesh Kawle , Aryan Kamble, Disha Hatnagar , Pranali Wanjari, Dr. Aparna Gale. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.