How do audit trails support AI governance?

• A. They document actions, data access, model changes, and decision points for accountability and audits.
• B. They track quarterly revenue.
• C. They determine salary bands.
• D. They store training data for all models.

Answer: (A)

Audit trails provide a recorded history of actions around data and models, capturing who accessed data, what changes were made to models, when those changes occurred, and why certain decisions were triggered. This traceability is crucial for AI governance because it creates accountability and enables audits. With a complete trail, you can verify that policies, access controls, privacy requirements, and risk mitigations were followed, and you can demonstrate compliance to regulators and stakeholders.

In practice, audit trails support data lineage and model provenance, showing how data flows from collection through training to deployment, and documenting model versions and the decision points that led to outputs. They also enable root-cause analysis after incidents, help with change management, and improve reproducibility of model behavior over time. By providing a tamper-evident record of actions and decisions, audit trails make governance processes transparent and auditable.

Other options don’t align with governance needs: tracking quarterly revenue is a business metric, not a governance mechanism; determining salary bands is HR-related; storing training data for all models concerns data storage and privacy, but without the accompanying logged actions, access, and changes, it doesn’t support the governance and accountability purpose of an audit trail.