The AI search retrieves an obsolete policy. How can we prevent this?
Stale knowledge retrievalGive documents clear ownership, effective dates, version status, and access rules. Ensure superseded material is excluded or clearly labeled for the relevant task, and test retrieval after updates. Search relevance cannot compensate for an unmanaged source collection.
What to do now
- Identify authoritative active sources
- Mark superseded documents and effective dates
- Refresh the retrieval index with access rules
- Test current and historical policy questions
When to escalate
Outdated guidance could authorize an incorrect action.
The AI service owner and relevant authorized reviewer
Example
A procurement answer points to the current approval threshold rather than a retired policy copy.
Mistake to avoid
Adding more documents while leaving old versions indistinguishable.
Prevent a repeat
Re-evaluate after changes to models, prompts, tools or source data; keep important decisions accountable to people.
Suggested ApprovedFlow setup
- Tasks
- Documents
- Approvals
Track evaluation and review tasks, and keep approved evidence with the work; do not claim the platform provides an unverified AI agent or automated diagnosis. See the platform
Related cases
- A colleague pasted confidential project documents into an unapproved AI service. What now?Unapproved information disclosure
- An uploaded document tells our AI to ignore its instructions. What should happen?Untrusted content attempting instruction override
- What should happen when a dependency is no longer needed?Stale dependency commitment
- Our estimates use outdated rates. What should we update?Stale cost assumptions