AI Safety & Responsible AI
Auditability
The ability to review an AI system's inputs, decisions, changes, and outputs after the fact.
Example
Logs and version records let an auditor reconstruct how an AI decision occurred.
Why it matters
It makes later investigation possible when someone questions a result or change.
You'll hear it
“Can we trace which version and records produced this decision?”
Can you control it?
Not directly
No, not directly. Auditability describes a learned property, observed result, or broader concept rather than a normal user setting. You may influence it indirectly by changing the tool, data, prompt, or workflow.
Common questions
- Does having logs guarantee a useful audit?
- No. Missing details, inconsistent records or inaccessible systems can still prevent reconstruction.
- Can too much logging create privacy problems?
- Yes. Storing unnecessary personal or confidential material can create additional exposure.
- Can a process be auditable without being fully explainable?
- Yes. Records may show what happened even when the internal reasoning is difficult to interpret.