Expanding Public Access to Understanding Data: A Case Study of Leveraging Generative AI for India Policy Insights Dashboard.

Abstract

Interactive dashboards have become central tools for visualizing policy-relevant data, but they remain limited by rigid structures and technical complexity. Recent advances in large language model (LLM)-powered conversational AI offer a way to address these challenges, as LLMs can interpret unstructured natural language inputs and translate them into structured functions. This proof-of-concept study introduces a Generative AI (GenAI)-powered chatbot that integrates LLMs with an online dashboard, demonstrated through the India Policy Insights (IPI) dashboard. The system translates conversational queries into structured functions, retrieves validated outputs from a spatially enabled database, and presents results as text, charts, and maps. We implemented 13 representative functions spanning spatial, temporal, composite, classification, and constraint-based analyses. Results show that spatial, classification, and constraint-based functions achieved consistently high accuracy due to explicit parameters, while composite multi-indicator functions posed greater challenges. The findings demonstrate the potential of GenAI-powered interactive dashboards to support language-driven interaction and broaden accessibility for diverse user groups, regardless of technological literacy. Beyond the case study, the proposed framework provides a design pathway for developing GenAI-powered dashboards that democratize access to spatiotemporal data, enhance evidence-based policymaking, and help bridge persistent gaps between data and actionable insights.