Year: 2024
Type: article
Abstract: Data governance -- the policies, processes, and technical controls that determine how data is collected, stored, used,shared, and deleted -- has become both more critical and more complex in the age of AI. AI systems consume data atunprecedented scale, require continuous data pipelines for retraining and monitoring, generate new data through theiroutputs (including synthetic data), and create novel privacy risks through memorisation and inference attacks.Simultaneously, AI can substantially automate data governance tasks that currently require manual effort: datacataloguing, lineage tracking, quality assessment, anomaly detection, and privacy compliance checking. This studyevaluates eight AI-driven data governance tools across five governance functions: automated data cataloguing andmetadata management, data quality monitoring and anomaly detection, privacy risk assessment (PII detection,re-identification risk), data lineage tracking, and GDPR/EU AI Act compliance checking. Tools evaluated includeLLM-based schema annotators (GPT-4-Catalog, DITTO), graph-based lineage trackers (DataLineageGNN), ML-baseddata quality monitors (GreatExpectations-ML, Anomalo), PII detection systems (Microsoft Presidio, AmazonComprehend), and an integrated governance platform (DataHub-AI). Evaluation used enterprise data environments from8 European organisations across four sectors. GPT-4-Catalog achieves 88.4% accuracy on metadata annotation across14,840 schema elements. DataLineageGNN tracks 94.2% of data transformation lineage vs. 68.4% for rule-basedbaselines. Anomalo detects 84.6% of data quality issues 2.4 days before they impact downstream pipelines. Agovernance maturity model with AI integration levels is proposed
Source: Bio-QI Journal
Authors: Lea Nowak, Anna Hansen, Laura Dubois
Language: English
Topic: Data Quality and Management
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W7203676751