Don't just see the number. Keep the evidence.
Keep a permanent record of AI responses, evidence, citations, and truth-record versions so you can trace what AI said about your brand over time.
A dashboard can tell you that an AI response was wrong.
The Response Audit Ledger lets you go back and see exactly what was said, when it was captured, which model produced it, and which version of your verified facts was used to evaluate it.
For teams that need a record of how AI represents their brand over time, that evidence matters.
Every AI response becomes part of the record
Each entry in the Response Audit Ledger captures the context behind an AI response, including:
- The exact response text
- AI model version
- Date and time
- Version of your truth record that was active
- Full evidence trail behind the verdict
- Citations associated with the response
So when you need to understand what happened, you don't have to rely on a dashboard number or remember what an AI model said weeks ago.
You have the underlying record.
How tamper-evidence works
Each captured response is stored with a hash, which acts as a deterministic fingerprint of the response text.
If the response text changes, its hash changes too.
This provides a way to detect whether the stored response has been altered. It is a tamper-evident record, not a claim of cryptographic immutability or blockchain-style security.
The result is a clearer evidence trail for the AI responses you're tracking.
Append-only records with long-term history
The Response Audit Ledger is designed as an append-only record.
Individual scan results cannot be deleted. Deleting the entire brand from Brand Hub removes its associated results.
Your retention period depends on your plan:
- Growth: 12 months
- Agency: 24 months
Growth and Agency also include PDF evidence export, so you can preserve and share the evidence outside AIVIS.
Growth and Agency include Manual Review, allowing you to flag or override a verdict when your review provides additional context.
Turn an AI response into evidence you can trace
When an AI model gets your brand wrong, you should be able to answer:
- What did it say?
- When did it say it?
- Which model produced it?
- Which version of our verified facts was used?
- What evidence supported the verdict?
The Response Audit Ledger keeps that context together.
Read the full Audit Ledger documentation →
Keep the evidence behind your AI visibility
Monitor what AI says about your brand today and keep a record you can return to tomorrow.
Response Audit Ledger is available on Growth and Agency plans.