Master Data Quality Challenges as a Barrier to AI Integration in Large-Scale Enterprise Systems
TL;DR
The article identifies three results: master data defects move through reporting and AI pipelines, S/4HANA transformations expose tolerated legacy errors, and AI readiness requires governed data quality gates before model deployment.
Abstract
The study examines master data quality as a constraint on artificial intelligence integration in large enterprise systems. Corporate AI projects often begin with model choice, automation scenarios, and platform design, while weak Business Partner and Material Master records remain inside ERP, SAP S/4HANA, BW, and analytical layers. The article shifts attention to that data foundation. Its aim is to explain how duplicated, incomplete, outdated, and poorly mapped master data restricts reliable AI use during ERP modernization. The review uses ten sources published from 2021 to 2025 and combines master data management, governance maturity, data-centric AI, enterprise AI adoption, and SAP migration literature. Comparative source analysis, conceptual synthesis, and typologization support the argument. The article identifies three results: master data defects move through reporting and AI pipelines, S/4HANA transformations expose tolerated legacy errors, and AI readiness requires governed data quality gates before model deployment. The proposed logic supports ERP, data governance, and AI implementation teams.