Argues that data governance is routinely treated as an afterthought in AI projects despite every AI system being, at its core, a data system.

Grounds the framework in a fictional online fashion retailer, ThreadLine Fashion, commissioning a customer churn prediction model from a third-party vendor.

Covers the full data governance lifecycle, including data quality, lineage and provenance, privacy, access control, bias and lifecycle management.

Frames data governance as a shared organisational responsibility that cannot be delegated to the AI vendor, since the organisation that owns the data owns the risk.

Aligns with ISO 42001, the ISO/IEC 5259 data quality series, ISO 27001, the EU AI Act and the NIST AI Risk Management Framework, and includes a maturity scorecard.