Claims processing · Production
ALFA and Shift Technology's Force detect organized claims fraud across French auto insurers
The problem
Organized fraud was impacting multiple French auto insurance companies simultaneously, with no cross-insurer mechanism to detect or coordinate against it collectively.
Workflow diagram · grounded in source
1
Combined member claims submitted
trigger
“By combining data from ALFA's participating members”
2
AI claims analysis against fraud scenarios
ai_action
“Force runs the claims data against a set of core fraud scenarios through an AI claims analysis engine”
3
Suspicious claims routed to ALFA
routing
“Claims that Force identifies as potentially fraudulent are sent to ALFA for further evaluation”
4
ALFA evaluates alert relevance
human_review
“If an alert is considered relevant”
5
Alert transmitted to member SIUs
output
“ALFA transmits it to member organizations as indication of organized fraud with supplemental information to aid the SIU investigation”
Reported outcome
Force met or exceeded ALFA's requirements and is now highly respected by similar organizations globally.
Reported metrics
French auto insurance claims analyzedover 30%
Reported stack
Force
Source
https://www.shift-technology.com/resources/case-studies/customer-stories/shift-alfa-join-forces-against-organized-fraud
Read source ↗Frequently asked questions
What did this team achieve with this AI workflow?
Force met or exceeded ALFA's requirements and is now highly respected by similar organizations globally.
What tools did this team use?
Force.
What results were reported?
French auto insurance claims analyzed: over 30% (source-reported, not independently verified).
How is this claims processing AI workflow structured?
Combined member claims submitted → AI claims analysis against fraud scenarios → Suspicious claims routed to ALFA → ALFA evaluates alert relevance → Alert transmitted to member SIUs.