Reviewing the Implementation Barriers of AI-Driven Data Governance Frameworks in Modern Enterprises

Abstract

Enterprises seek AI data governance to automate compliance with regulations, governance, and high-quality data. Striking a balance between realizing this promise and implementing production-grade, sustainable practices remains a challenge for many businesses. This paper examines and analyzes the technical, organizational, legal, and cultural factors that hinder the adoption of AI data governance frameworks in contemporary business enterprises. Data was analyzed from a cross-sectional survey of data engineers, data governance leads, compliance officers, and AI/ML practitioners, comprising 100 individuals. A descriptive analysis (in terms of frequency and percentage) of current practices, perceived obstacles, and areas for investment was conducted. The core findings indicate that the primary obstacles affecting AI data governance in most enterprises include gaps in data quality and lineage, siloed functional collaboration, immature Machine Learning Operations (MLOps), and tooling, as well as legal and privacy issues related to sensitive data and concerns about multi-role capacity. The research aims to increase the success rate of AI-driven governance initiatives by applying a practical, step-by-step approach based on the evidence collected.