Claims processing · Production
Generali France uses Shift Technology Force to increase fraud detection and reduce false positives on property claims
The problem
Generali France needed to increase its detection rate on fraudulent property claims while processing a high volume of annual claims.
Workflow diagram · grounded in source
1
Claim filed
trigger
“From the moment a claim is filed, Force is capable of combining thousands of data variables”
2
Multi-variable data analysis
ai_action
“Force is capable of combining thousands of data variables and identifying irregularities among the claim information”
3
Real-time anomaly detection
ai_action
“detect anomalies in claims in real time to find fraud faster”
4
Investigator review
human_review
“give investigators a head start on verifying or eliminating fraud prior to claims payment”
Reported outcome
Force significantly reduced false positives, speeding up claims processing and saving time for both the insured and insurer, while enabling real-time anomaly detection to find fraud faster.
Reported metrics
False positive ratesignificantly reduce the number of false positive
Claims processing periodspeeding up the processing period of claims
Time saved for insured and insurersaving time for both the insured and insurer
Annual claims processedover 100,000
Reported stack
Force
Source
https://www.shift-technology.com/resources/case-studies/customer-stories/generali-france-to-increase-detection-rate
Read source ↗Frequently asked questions
What did this team achieve with this AI workflow?
Force significantly reduced false positives, speeding up claims processing and saving time for both the insured and insurer, while enabling real-time anomaly detection to find fraud faster.
What tools did this team use?
Force.
What results were reported?
False positive rate: significantly reduce the number of false positive; Claims processing period: speeding up the processing period of claims; Time saved for insured and insurer: saving time for both the insured and insurer; Annual claims processed: over 100,000 (source-reported, not independently verified).
How is this claims processing AI workflow structured?
Claim filed → Multi-variable data analysis → Real-time anomaly detection → Investigator review.