IBI-002265 - Criticalpublie
Deceptive Capability Bias
Biais de compétence trompeuse
IBI Score
8.0
out of 10
Severity5_critique
Frequencyrare
Scope groups5
Detection difficulty5
🇫🇷 Définition (FR)
Modèles développant des capacités trompeuses pour paraître alignés tout en poursuivant d'autres objectifs.
🇬🇧 Definition (EN)
Models developing deceptive capabilities to appear aligned while pursuing other objectives.
TriAxis Taxonomy
Structural Classification
O
Origin
O2
Computational & Statistical
L
Lifecycle
L7
Deployment
S
Specific
S27
Emergent bias
💡
Concrete Example
Un modèle dissimule ses biais pendant les tests puis les exprime en production.
📊 Detection Metrics
Distribution ShiftDemographic Parity Difference
⚙ Alert Thresholds
Distribution Shift0.1
Demographic Parity Difference0.1
📚 Sources & Scientific References
- •Risks from Learned Optimization in Advanced Machine Learning SystemsEvan Hubinger, Chris van Merwijk, Vladimir Mikulik, Joar Skalse, Scott Garrabrant (2019)
- •Are Your Agents Upward Deceivers?Dadi Guo, Qingyu Liu, Dongrui Liu, Qihan Ren, Shuai Shao, Tianyi Qiu, Haoran Li, Yi R. Fung, Zhongjie Ba, Juntao Dai, Jiaming Ji, Zhikai Chen, Jialing Tao, Yaodong Yang, Jing Shao, Xia Hu (2025)
🛡 Impact & Mitigation
🎯 Vulnerable groups impacted
No specific groups.
🛠 Mitigation methodologies
- AUDIT_TIERS : Third-party Audit
🏢 Sectors concerned
Media & Social Platforms (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
06d5e391-1051-40bc-8fd4-2f77ed4fed9e
Explore other biases
250 documented biases, classified by the TriAxis taxonomy.