OECD AI Principles OECDAI-5: Data Governance, Training Data Quality, Privacy, and Bias Mitigation

Operate data governance underpinning trustworthy AI per OECD Principles. Data governance must address (a) training data quality and governance with documented sourcing + provenance + consent + licensing + curation + quality controls + (b) data bias assessment and mitigation across training + validation + testing + production data, (c) data provenance and lineage tracking with metadata management + chain of custody + reproducibility support, (d) privacy protection in AI training data per applicable privacy regimes (GDPR + state privacy laws + sector-specific) + including data minimisation + purpose limitation + lawful basis + data subject rights handling for training data + model output containing training data + (e) data retention for AI models including training data + model snapshots + audit trails + inference logs per regulatory + investigative + governance need, (f) bias detection and mitigation including evaluation against fairness metrics appropriate to use case + intervention at data + model + output stages + ongoing monitoring for emergent bias + (g) cross-border data flow for AI training and inference per applicable regulation + (h) special handling for sensitive categories (health + biometric + children + protected class).