Investigate a security concern that AI Control Tower has detected on a managed AI asset, take the appropriate remediation action, and document the resolution.
Before you begin
Role required: sn_ai_asset_mgmt.ai_asset_owner or sn_ai_governance.ai_steward
Procedure
-
Navigate to .
-
Select the Security tasks sub-tab.
-
Select the security task that you want to resolve.
The task opens in a side panel with the affected asset, the type of
security concern that triggered the task, the supporting evidence, and any
recommended remediation actions.
-
Investigate the security concern.
Review the evidence in the task, examine the asset's security posture on
the asset record's Security tab, and determine
whether the concern is a true positive that needs remediation or a false
positive that can be dismissed.
-
Take the remediation action that addresses the security concern.
Remediation actions vary by the type of concern. For example, a prompt injection task might require updating the agent's guardrails, while a dormant agent task might require retiring the agent or rotating its
credentials.
-
Document the resolution and update the task status.
Record the action you took and the outcome in the task's resolution notes
so the security audit trail is complete. Set the task status to
Completed when the work is done.
Result
The resolved security task no longer appears in the Assigned to
you view. The asset record's Security tab
reflects the updated security posture. If the underlying condition recurs, AI
Control Tower generates a new security task.