This paper explores how Large Language Models (LLMs) may support judicial activism in India by improving access to legal knowledge, facilitating rights-based research, and identifying patterns in case law across jurisdictions and languages. In a judicial system shaped by Public Interest Litigation (PIL) and facing challenges of scale, linguistic diversity, and doctrinal complexity, LLMs offer practical capabilities in semantic analysis, summarisation, and multilingual processing. Drawing on examples from the United Kingdom, the European Union, Australia, and New Zealand, as well as domestic initiatives like SUPACE and SUVAS, the paper argues for the careful integration of LLMs as assistive infrastructure that enhances—rather than replaces—human legal reasoning. Key concerns around accuracy, bias, transparency, and accountability are addressed through a proposed human-in-the-loop framework, multilingual model development, and institutional safeguards. The paper concludes that, if governed responsibly, LLMs can strengthen the judiciary’s capacity for timely, inclusive, and constitutionally grounded interventions.