What challenges did Wisconsin Department of Workforce Development face before AI implementation?

Unemployment claim volumes surged beyond the capacity of existing digital infrastructure. Adjudicators needed a secure way to validate claims, identify potentially fraudulent cases, and release backlogged payments quickly. Manual processing created risk of delays and human errors in recharge payment calculations.

What was the impact of the AI solution?

The agency could more quickly filter potentially fraudulent claims while continuing to process legitimate claims. Automation improved adjudicator efficiency and supported faster claim processing. Automatic calculations and processing of recharge amounts reduced human errors. The approach helped ensure benefits reached the right individuals more quickly.

Source

GCPEvidence: Low25/100

Wisconsin Department of Workforce Development fraud detection for unemployment claims using BigQuery ML and AutoML

Use case typeWorkflow automationUpdated Jun 13, 2026

The Wisconsin Department of Workforce Development needed to process unprecedented volumes of unemployment claims while separating potentially fraudulent or improper-payment cases from legitimate claims. SpringML and Google Cloud implemented an AI/ML fraud-detection framework that uses cloud data storage, machine learning, and dashboards to help adjudicators validate claims and prioritize urgent cases. The solution also automated recharge amount calculations to reduce human errors and speed processing.

Organization: Wisconsin Department of Workforce Development
Industry: Public Sector
Location: United States
Published: May 2021

Reported outcomes

Strategic outcomes

Risk & complianceValidated claims and flagged fraud fasterSpeed & agilityAccelerated claim processingRisk & complianceReduced human errors in calculationsCustomer experience & trustReached eligible recipients sooner

Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks

Customer: Wisconsin Department of Workforce Development
Provider: GCP
Maturity: Unknown
Linked source: Google Cloud Blog

No explicit deployment-stage evidence found.

Customer identity supportedSource describes one deploymentMaturity evidence evaluated

Primary read

Use case focus

Showing 3 of 4

1Fraud detection
2Claims processing
3Anomaly detection

Unemployment claim volumes surged beyond the capacity of existing digital infrastructure.
Adjudicators needed a secure way to validate claims, identify potentially fraudulent cases, and release backlogged payments quickly.
Manual processing created risk of delays and human errors in recharge payment calculations.

SpringML used Google Cloud Storage and BigQuery to store and manage claims data.
BigQuery Machine Learning and AutoML were used to build predictive models, risk scoring, anomaly detection, and fraud classification.
Looker and Data Studio dashboards helped leaders and adjudicators review claims and make faster decisions.
The framework included unsupervised machine learning and configurable rules that could be updated as fraud patterns changed over time.

Technologies

Google Cloud Storage BigQuery BigQuery Machine Learning AutoML Looker Data Studio

The agency could more quickly filter potentially fraudulent claims while continuing to process legitimate claims.
Automation improved adjudicator efficiency and supported faster claim processing.
Automatic calculations and processing of recharge amounts reduced human errors.
The approach helped ensure benefits reached the right individuals more quickly.

Sources & evidence1

Evidence: Low25/100Evidence strength