Health insurer's datamining model finds more irregular CMU-C files

The Cour des comptes found that data mining by France's national health insurer helped raise the irregularities found in CMU-C files selected for checks to 24% in 2018.

  • Not rated an organization

    The Cour gives the 2018 share of irregular files but no share from before the model, so the size of the gain cannot be established.

Depth runs from 1, a convenience or time saved, to 5, a life or a livelihood preserved.

Status
Reported benefit.
Evidence

foundstudiedreportedclaimed

An independent audit report supports the benefit.

The system’s part
Not isolated. The Cour describes the system only as data mining and gives no detail of its method or of whether it learned from data. The Cour credits the model with contributing to the higher rate but reports no rate without it.
Limits
Not yet checked.
Costs and harms
Not yet checked.
Later
Not yet checked.

Download data Cite Suggest a correction

Benefit type
Improved Task Accuracy
System
Data-mining model for after-the-fact checks of CMU-C files
Made by
Caisse nationale de l'assurance maladie (Cnam)
Used by
Caisses primaires d'assurance maladie (CPAM), all funds since June 2019
Where
France (national)
Stage
Expanding
Date
Kind
Aggregate result
Broad benefit
Analyzing Data or Information, Evaluating Information to Determine Compliance with Standards
Collections
Scrutiny and findings
Last revised
31 October 2026

Sources

  1. Audit report, by authors with no tie to the system.

    Cour des comptes. La lutte contre les fraudes aux prestations sociales. Communication à la commission des affaires sociales de l’Assemblée nationale, September 2020, p. 51.

    La diffusion progressive de ce dispositif à partir de 2018 a contribué à accroître la fréquence des irrégularités détectées sur les dossiers sélectionnés au contrôle (24 % en 2018, dont 9 % qualifiées de frauduleuses).

Details

AI system

Workflow

The model selected CMU-C files for after-the-fact checks by the local funds.

Follow-up

Conditions for replication

The local funds must check the files that the model selects.

Entry details

Sector

Health insurance (complementary cover for people on low incomes).

Purpose

The Cnam built the model to select CMU-C files for after-the-fact checks.

Not yet checked: model, release, provider, role, configuration, inputs and outputs, autonomy, human oversight, AI contribution, beneficiaries, reach, distribution and access, duration, every measurement question, every money question, ownership and control, dependencies, cross-border connections, geopolitical implications, replication attempts, availability, institutional decision, grounds, action and result, success criterion, every evidence question, observed or intended, alternative assessments, related entries, other identifiers, category mappings.

Cite this record

Peter H. Vartanian. (2026). Health insurer's datamining model finds more irregular CMU-C files (AIBD-0687). Artificial Intelligence Benefits Database, edition of 31 October 2026. https://petervartanian.xyz/aibd/records/AIBD-0687.html (archived at https://doi.org/10.7910/DVN/AADSII, version 4.0)

BibTeX Data Field definitions