---
sourceDocument: Zurich Enable AI
sourceDocumentLink: https://www.servicenow.com/docs/r/zurich/intelligent-experiences
Release :
- zurich
ft:locale :
- en-US
ft:publication_title :
- Zurich Enable AI
ft:clusterId :
- platai
bundleId :
- platai
workflow :
- Platform
---
# Approval assistance AI agent
# Platform Approval assistance AI agent {#ariaid-title1}
Release version: Zurich
Updated July 27, 2026
 2 minutes to read
Summarize
 Summarized using AI
This content was generated using new OpenAI-powered functionality. Results are provided on an as is basis and are not guaranteed to be accurate or complete.
## Summary of Platform Approval assistance AI agent
The Approval assistance AI agent streamlines the approval process within ServiceNow by allowing approvers to view and manage their pending approvals directly through ServiceNow Otto® in Virtual Agent.
Users can inquire about their pending approval requests and take actions such as approving or rejecting requests without navigating to specific pages.
The AI agent provides context-aware responses and follow-up questions to simplify decision-making.
Show full answer Show less
This AI agent accesses approval data from the **Approvals \[sysapprovalapprover\]** table, showing records assigned to the user for approval. It requires specific roles for access and data visibility, including **approveruser, snrequestapproverread,** and **sncinternal** , along with additional roles to read other approval-related tables (e.g., **snchangeread** for Change Requests).
## Key Features
* **Role-Based Access:** Ensures users see only approvals relevant to their assigned roles, with role masking controlling user access to the AI agent.
* **Context-Aware Interaction:** The agent asks clarifying questions and provides detailed approval information to aid decision-making.
* **Approval Actions:** Users can approve or reject requests directly within the AI agent interface, including adding comments when rejecting approvals.
* **Configurable Decision Support:** Administrators can configure display fields and Knowledge Base (KB) search fields to create a Gen AI checklist. This checklist helps the AI agent leverage KB articles and policies for informed approval assistance.
* **Cross-Scope Access:** For applications with scoped restrictions, cross-scope privileges must be granted to ServiceNow Otto Agents for Requester plugin to enable proper data access.
* **Integration with ServiceNow Otto and Virtual Agent:** The AI agent operates seamlessly in these interfaces for convenient access.
## Setup and Access
* The AI agent is installed with certain applications; customers should verify their entitlements.
* Administrators must assign the required roles to users and configure security controls to enable AI agent access and data visibility.
* To access and manage the AI agent, navigate to **All \> AI Agent Studio \> Create and manage**, then select the Approval assistance AI agent under the AI Agents tab.
## Practical Usage Examples
Users can interact with the AI agent using natural language queries such as:
* "Can you give me a list of pending approvals?"
* "What are the pending approvals for time off requests?"
* "Show me a list of pending approval requests by priority."
* "Give me details about my approval request."
After receiving approval details, users can command the AI agent to approve or reject requests, streamlining the approval workflow.
The Approval assistance AI agent is an AI agent that enables you to see your list of pending approvals, and the details about your pending approvals. You can then approve or reject requests and tickets from ServiceNow Otto® in Virtual Agent.
## Approval assistance AI agent overview {#platform-approval-aia__section_kqz_3ry_1hc}
Roles required: approver_user, sn_request_approver_read, snc_internal
Note:
Add the necessary roles to enable reading of other tables whose records go through the approval process. For example, you can add the sn_change_read role to read Change Request records.
The benefit of using the Approval assistance AI agent is that you don't need to navigate to a specific page to approve your tickets. You can ask the AI agent about your pending approval requests and then tell the AI agent to
approve or reject those approvals. The AI agent will ask follow-up questions and offer context-aware responses to simplify your experience.
Ask you administrator to configure the display fields and the knowledge base (KB) search fields to generate a Gen AI checklist to assist the Approval assistance AI agent in making targeted decisions. The checklist uses KB articles and policies to assist the Approval assistance AI agent in decision making. The checklist fetches information from knowledge base articles about specific requester approval tickets. An approval_admin and admin role are required to configure the Approval assistance AI agent. For more information, see [Configure Service Portal Approval Configuration record](https://www.servicenow.com/docs/access?context=configure-approval-assistance-ai-agent&version=zurich&pubname=zurich-platform-user-interface&ft:locale=en-US).
Note:
Provide cross-scope privileges to the ServiceNow Otto Agents for Requester plugin for tables whose records are restricted within the scope of an application. For more information, see [Define cross-scope access to an application resource](https://www.servicenow.com/docs/access?context=set-RCA-level&version=zurich&pubname=zurich-application-development&ft:locale=en-US).
## Prerequisites and setup {#platform-approval-aia__section_apx_t2z_1hc}
This AI agent is installed with certain applications. Check your entitlements for details.
The Approval assistance AI agent displays data from the Approvals \[sysapproval_approver\] table. If the user has been assigned to approve a request, the approval record is shown.
## Role masking {#platform-approval-aia__section_nf3_dck_jhc}
AI agents use [role masking](https://www.servicenow.com/docs/Y5Ut9VQBSiXhQFMyYJC_fw "Role masking for AI agents and agentic workflows helps users enhance security by enabling them to limit their roles during tool execution and verify that AI agents run with least-access privileges.") to determine which users can access them. Ones installed with your applications have specific roles that come included with the application. If you select Users with specific roles for user access, you must configure the security controls to include these roles. For the instructions to change the security controls, see [Define security controls for an AI agent](https://www.servicenow.com/docs/dXvvVFPByrXRUHxE_roitw "In the guided setup for an AI agent, define security controls for who can access the AI agent and what data the AI agent has access to.").
In the data access settings, you must also add the necessary roles to enable reading of the tables for the records you want to evaluate for readiness. For example, you can add the itil role to the AI agent's list of approved roles
so that it can access Incident records.
## Accessing the Approval assistance AI agent {#platform-approval-aia__section_x3k_vmz_1hc}
To access the AI agent:
1. Navigate to AllAI Agent StudioCreate and manage.
2. Go to the AI Agents tab.
3. Select Approval assistance AI agent.
{#platform-approval-aia__ol_y3k_vmz_1hc}
## Sample
utterances {#platform-approval-aia__section_x4x_kw1_bhc}
After the agent has been activated in AI Agent Studio, enter phrases such as the following or similar queries to run the AI Agent in Virtual Agent and the ServiceNow Otto panel.
* Can you give me a list of pending approvals?
* What are the pending approvals for time off requests?
* Show me a list of pending approval requests by priority.
* Give me details about my approval request?
{#platform-approval-aia__ul_y4x_kw1_bhc}
## Performing approval action on an approval record {#platform-approval-aia__section_d1t_ddh_bhc}
The Approval assistance AI agent shows approval requests that require an action on your part. After the approval information is provided by the agent, select Approve or Reject for a specific approval record, or type Approve or Reject. If you reject an approval request, then add a comment about the rejection reason. Figure 1. Approve or reject an approval record
*[\>]: and then