---
title: Summarize fraudulent claim detection using agents
description: Automatically detect fraudulent claims by identifying duplicates, similar claim patterns, reused dealer images, mismatched or incomplete supporting documents, usage inconsistencies, and abnormal claim spikes.
locale: en-US
canonical_url: https://www.servicenow.com/docs/r/manufacturing/mco-summarize-fraudulent-claim-detection.html
release: brazil
topic_type: task
last_updated: "2026-09-10"
reading_time_minutes: 1
keywords: [warranty claim summarization]
breadcrumb: [Repair claim, MCO workspace, Use, Manufacturing Commercial Operations]
---

# Summarize fraudulent claim detection using agents

Automatically detect fraudulent claims by identifying duplicates, similar claim patterns, reused dealer images, mismatched or incomplete supporting documents, usage inconsistencies, and abnormal claim spikes.

## Before you begin

Role required: admin

Enable the skill before use. It is inactive by default:

1.  Navigate to **All** > **ServiceNow Otto Admin** > **Skills**.
2.  Select **Customer**.
3.  Select **MCO**.
4.  Activate the skill **Detect Claim Anomaly**.

## About this task

The App Anomaly Detection Rule \[com.sn\_anomaly\_detect\] framework identifies fraudulent claims. The Agents portal displays fraudulent repair claims submitted from the dealer portal using these rules.

-   Similar claim: Identifies the similar claims submitted for the same customer across multiple assets or vehicles.
-   Duplicate claims: Automatically identifies claims that match previously processed claims for the same vehicle, part, or repair.
-   Image reuse: Detects claims that reference the same picture as a prior submission.
-   Configurable detection parameters: Allows administrators to tune detection sensitivity and time windows to match their operational policies.
-   Product usage mismatch detection: Compares product usage data between the current claim and historical claim records to flag inconsistencies.
-   Dealer claim spike: Surfaces similar claims submitted by the same dealer within a configurable time window.
-   Document mismatch: Detects claims with mismatched or incomplete supporting documents.
-   Threshold based approval: Detects repair claims that exceed threshold limits that have been specified for the causal parts.
-   Inflated claim: Detects if the claim amount is significantly greater than the average claim amount for similar claims.
-   Cross validation: Compares a repair claim's charges against the pre-authorization request or recall campaign it originates from. Claims that fall outside the configured tolerance are flagged for manual review.

Use case 1: Dealer submits repair claim with image.

To submit the report in the Dealer portal, see [Submit a repair claim for warranty](https://raw.githubusercontent.com/ServiceNow/ServiceNowDocs/brazil/markdown/manufacturing/mco-submit-repair-claim.md).

The repair claim appears in the Agents portal with Reject status.

## Procedure

1.  Navigate to **Workspaces** > **CRM Workspace** > **Lists** > **Repair**.

2.  Select the repair claim record that you want to review.

    **Note:** Choose the record that has the **Potential action** as Reject/Suggested.

3.  Select **Review & approve**, to view the claim analysis generated by ServiceNow Otto \[Omitted image "icon-ai-sparkle.png"\] Alt text:.


**Parent Topic:**[Repair claim](https://raw.githubusercontent.com/ServiceNow/ServiceNowDocs/brazil/markdown/manufacturing/mco-use-repair-claim.md)