Prerequisite: None

This session will include a moderated Q&A featuring questions from the live audience.

The rapid adoption of AI-driven applications is forcing organizations to reconsider many of the assumptions that have traditionally guided enterprise data architects in supporting end-user analytics capabilities. Conventional analytics environments are engineered to support conventional analyses: data aggregation, reporting, and visualizations for decision support. Modern AI techniques, including natural language processing (NLP), large language models (LLMs), and transformer-based architectures require significantly greater semantic context and a deeper understanding of relationships embedded across both structured and unstructured information artifacts.

TDWI research shows that this shift introduces new integration challenges, especially when integrating AI-driven components such as intelligent chatbots and prompt-driven visualizations providing consistent and trustworthy intelligence. At the same time, many of the persistent issues that have undermined confidence in analytics for decades—poor data quality, inconsistent business definitions, limited metadata visibility, inadequate lineage, and fragmented governance—have not disappeared, but instead become more consequential in the context of generative AI. The effectiveness of AI initiatives increasingly depends less on model sophistication alone and more on the organization’s ability to establish trusted, semantically coherent, and well-governed data foundations.

In this session, TDWI research fellow David Loshin will examine the implications of these changes and provide practical perspectives on where organizations should focus their investments to support both trustworthy analytics and scalable AI enablement.