Integration of Large Language Models with Power BI for Automatic Insight Generation: Integración de Modelos de Lenguaje de Gran Escala con Power BI para la Generación Automática de Insights

Abstract

Introduction: The growing availability of corporate data has increased the need for tools that facilitate analysis without requiring advanced technical expertise, fostering the integration of generative artificial intelligence into business intelligence platforms such as Power BI.
Objective: To analyze the integration of large language models with Power BI to enable augmented analytics capabilities focused on automatic insight generation.
Method: A conceptual review and analytical approach was conducted on business intelligence, the Transformer architecture, and automatic insight generation, complemented by the proposal of a five-layer architecture and the review of implementation patterns using Power BI Copilot, Azure OpenAI, and custom APIs.
Results: Key capabilities were identified, including natural language-to-query translation, automatic DAX code generation, explanatory narrative creation, and conversational interaction with data, as well as applications in retail, human resources, and supply chain.
Conclusions: The integration of large language models with Power BI represents a significant step toward democratizing data analysis by enabling non-specialized users to access insights in an agile, understandable, and effective manner.