IBI Score
7.0
out of 10
Severity4_grave
Frequencyoccasionnel
Scope groups4
Detection difficulty4
🇫🇷 Définition (FR)
Apparition de nouveaux sous-groupes en production non présents lors de l'entraînement.
🇬🇧 Definition (EN)
Appearance of new subgroups in production not present during training.
TriAxis Taxonomy
Structural Classification
O
Origin
O2
Computational & Statistical
L
Lifecycle
L7
Deployment
S
Specific
S15
Population bias
💡
Concrete Example
Un modèle médical entraîné en 2015 sous-performe pour les patients post-pandémie.
📊 Detection Metrics
Distribution ShiftDemographic Parity Difference
⚙ Alert Thresholds
Distribution Shift0.1
Demographic Parity Difference0.1
📚 Sources & Scientific References
- •
- •Popular sepsis prediction tool less accurate than claimedMichigan Medicine (communiqu�, contact Kelly Malcom) ; �tude : Andrew Wong, Karandeep Singh et al., JAMA Internal Medicine (2021)
🛡 Impact & Mitigation
⚠ Documented evidence (Studies & Incidents)
- Mod�le de pr�diction du sepsis d'Epic jug� peu performant lors d'une validation externe : Source : https://www.patientsafetysolutions.com/docs/July_2021_EPIC_Sepsis_Prediction_Tool_Falls_Short.htm
Impact:
🎯 Vulnerable groups impacted
No specific groups.
🛠 Mitigation methodologies
- PSI : Population Stability Index
- AUDIT_TIERS : Third-party Audit
🏢 Sectors concerned
Healthcare & Medicine (Risk: /5)
⚖ Applicable regulations
🌍 International
- •ISO_42001 : ISO/IEC 42001:2023
- •UNESCO_2021 : UNESCO - Recommandation sur l'éthique de l'IA
- •OECD_AI_PRINCIPLES : OCDE - Principes IA
- •COE_AI_CONVENTION_INT : Conseil de l'Europe - Convention IA
📋 Metadata
Contributor
Mohamadou SIDIBE
Institution
AI BIAIS INITIATIVE
Version
1
Created on
October 14, 2026
Updated on
November 4, 2026
UUID
15a1c4a0-b4ab-4551-804b-66b6dda49115
Explore other biases
250 documented biases, classified by the TriAxis taxonomy.