A Business Intelligence-Based Data Governance Model to Reduce the Risk of AI Implementation Failure in Organizations

The implementation of Artificial Intelligence (AI) in organizations is increasing along with the development of digital transformation and the need for data-driven decision-making. However, various studies show that many AI projects fail due to poor data quality, lack of data integration, and weak data governance within the organization. In this context, Business Intelligence (BI) has the potential to support systematic data management through data integration, analytics, and information visualization. Therefore, a data governance model integrated with Business Intelligence is needed to improve the quality of data management and reduce the risk of AI implementation failure. This study aims to develop a Business Intelligence-based Data Governance model that can reduce the risk of Artificial Intelligence implementation failure in organizations and increase the effectiveness of data-driven decision-making. This study uses a quantitative approach with an explanatory research method. Research data were obtained by distributing questionnaires to respondents involved in data management and the implementation of organizational analytical systems. Data analysis was conducted using the Structural Equation Modeling (SEM) method with the Partial Least Squares (PLS) approach to examine the relationship between data governance variables, Business Intelligence capability, and AI implementation risk. The results of this study indicate that data governance significantly impacts Business Intelligence capabilities, which in turn contributes to reducing the risk of AI implementation failure within organizations. Furthermore, data quality is shown to be a crucial mediating factor linking data governance to successful AI implementation. This study produces a Business Intelligence-based Data Governance conceptual model that can be used as a framework for organizational data management to support more effective AI implementation.