---
sourceDocument: Australia Security Management
sourceDocumentLink: https://servicenow-prod.fluidtopics.net/r/security-management

 Release :

    - australia

ft:locale :

    - en-US

ft:publication_title :

    - Australia Security Management

ft:clusterId :

    - security

bundleId :

    - security

workflow :

    - Technology


---

# Use the AI guardrails helper skill

# Use the AI guardrails helper skill {#ariaid-title1}

* Release version: Australia
* 
* Updated August 3, 2026
* 
* ![](https://www.servicenow.com/docs/portal-asset/ico-clock) 1 minute to read

This AI skill identifies finding types and helps you understand guardrails that are already mapped to findings and why they were selected by the skill for mapping. This information helps you determine any findings that might
be mitigated and deferred for later review and remediation.

## Before you begin

The ServiceNow Otto® panel must be activated. For more information, see [Activate the ServiceNow Otto panel standard chat](https://www.servicenow.com/docs/access?context=activate-now-assist-panel&version=australia&pubname=australia-intelligent-experiences&ft:locale=en-US).

Roles required: sn_vul.vulnerability_admin or sn_vul.vulnerability_analyst admin

## Procedure

1. Navigate to WorkspacesSecurity Exposure ManagementAI Exposures.
2. Select the AI validation findings tab.  
   The AI guardrails helper panel is displayed.  
   Note:  
   This panel is displayed only if you have installed the ServiceNow Otto for Unified Security Exposure Management application.
3. If you're using the application for the first time, select Detect guardrails.
4. Respond to the prompt that asks you if the AI guardrails helper should run periodically.  
   * Yes --- Run the job on the scheduled time and frequency. You can choose how often the job runs.
   * No, only manually (on-demand) --- Run the skill manually when needed.

   {#ai-security-exposure-use-aiskill__ul_hp4_4qw_w3c}

   To view these options after you initially set them, select the menu icon ![]()
5. Select Confirm and run.  
   Run time scales with the number of newly imported findings. The first run may take a few moments to display results. Subsequent runs process only the delta findings. You can activate a notification to inform you when
   the job completes.  
   * N (number of findings) for deferral --- The number of findings that can use detected matching guardrails. These findings can be deferred, because the risks can be mitigated by the guardrails.
   * Mitigated findings by threat category, for example:
     * Social Division and Polarization
     * Violence and Public Safety
     * Hate Speech
     * Harassment
     {#ai-security-exposure-use-aiskill__ul_mhv_mrw_w3c}
   {#ai-security-exposure-use-aiskill__ul_pn2_krw_w3c}

   Select a value from the categories to open a list of findings by category (AIVUL).

   You can now use the Guardrail Detector agentic workflow to create exception rules to automatically defer findings that match your criteria.
{#ai-security-exposure-use-aiskill__steps_tdx_vgk_hjc}

*[\>]: and then