IBI-000435 - Criticalpublie
Multi-agent Amplification Bias
Biais d'amplification multi-agents
IBI Score
9.0
out of 10
Severity5_critique
Frequencyfrequent
Scope groups5
Detection difficulty5
🇫🇷 Définition (FR)
Amplification des biais par interaction entre plusieurs agents IA autonomes.
🇬🇧 Definition (EN)
Amplification of bias through interaction between multiple autonomous AI agents.
TriAxis Taxonomy
Structural Classification
O
Origin
O2
Computational & Statistical
L
Lifecycle
L7
Deployment
S
Specific
S02
Feedback loop bias
💡
Concrete Example
Des agents négociateurs reproduisent et amplifient des biais de caste dans les transactions.
📊 Detection Metrics
Feedback Loop RatioDemographic Parity Difference
⚙ Alert Thresholds
Feedback Loop Ratio0.15
Demographic Parity Difference0.1
📚 Sources & Scientific References
- •Multi-Agent Risks from Advanced AILewis Hammond, Alan Chan, Jesse Clifton, Jason Hoelscher-Obermaier, Akbir Khan et al. (2025)
- •Judging with Many Minds: Do More Perspectives Mean Less Prejudice? On Bias Amplifications and Resistance in Multi-Agent Based LLM-as-JudgeChiyu Ma, Enpei Zhang, Yilun Zhao, Wenjun Liu, Yaning Jia, Peijun Qing, Lin Shi, Arman Cohan, Yujun Yan, Soroush Vosoughi (2025)
🛡 Impact & Mitigation
🎯 Vulnerable groups impacted
No specific groups.
🛠 Mitigation methodologies
- AUDIT_TIERS : Third-party Audit
🏢 Sectors concerned
Finance & Banking (Risk: /5)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
8fd57f88-fecc-41e8-a5fe-d9c1b1fc6525
Explore other biases
250 documented biases, classified by the TriAxis taxonomy.