AUBIQ: A Generative AI-Powered Framework for Automating Business Intelligence Requirements in Resource-Constrained Enterprises

In today’s ever-changing data-driven environment, obtaining requirements for business intelligence (BI) systems remains a major challenge. This pain point is particularly prominent in small and medium-sized enterprises with limited resources: weak technical teams, high external communication costs and long response cycles make traditional manual specification processes more difficult to maintain. This paper proposes a new AI-driven codeless framework AUBIQ, which uses semantic search and large language models (LLM) to automate and accelerate the specification of BI system requirements. The system converts user natural language input into outputs such as prototype analysis code, descriptions, and data dependencies through a conversational interface, and can generate detailed test case reports with visual assistance, thereby streamlining the requirements acquisition process. In addition, AUBIQ introduces a user feedback mechanism to continuously improve BI reports and system design, showing practical application potential for accelerating data-driven decision-making. This paper also explores the broad prospects of generative artificial intelligence in changing the development of BI systems and explains its role in large-scale evolving data engineering practices.