How do auditing capabilities support accountability in AI systems?

• A. They reveal data provenance and decision rationales.
• B. They reduce data storage needs.
• C. They accelerate training times.
• D. They make audits unnecessary.

Answer: (A)

Auditing capabilities enable accountability by providing traceability of how AI systems operate. They capture data provenance—where inputs come from, how they were prepared, and which data was used—and document decision rationales, including why a particular output or action was chosen. This kind of transparent logging, along with model versions, configurations, and time-stamped events, lets someone verify that processes followed policies, standards, and regulatory requirements. It also supports root cause analysis after incidents, bias detection, and governance review, because you can reconstruct the exact steps that led to a result and explain it to stakeholders.

The other options don’t directly support accountability. Reducing data storage helps efficiency but not the ability to explain or justify decisions. Accelerating training times improves performance but doesn’t provide traces or rationales for outputs. Making audits unnecessary defeats the purpose of accountability in the first place.