Unfair capability distribution
AI Risk"Performing worse for some groups than others in a way that harms the worse-off group"
Overview
A source-backed snapshot of this AI risk.
Techniques0Attack methods connected to this risk.
Mitigations0Defenses that may help with related attacks.
Records2Source records unified into this concept.
Risk profile
How the MIT AI Risk Repository categorizes this risk.
Domain1. Discrimination & Toxicity
Subdomain1.3 > Unequal performance across groups
Entity2 - AI
Intent2 - Unintentional
Timing2 - Post-deployment
CategoryRepresentation & Toxicity Harms; Representation and Toxicity
SubcategoryUnfair capability distribution
Merged risk records
Source records unified into this canonical risk concept.
2 recordsView all →
MITRISK-Weidinger2023-18.01.02 - Unfair capability distribution
"Performing worse for some groups than others in a way that harms the worse-off group"
MITRISK-Li2025-66.06.03 - Unfair capability distribution
"Performing worse for some groups than others in a way that harms the worse-off group"
Mitigations
Defenses that may help with related attacks.
No propagated mitigations. No defense is available through the connected attack methods.
Source evidence
Research source for this risk, when available.
Included resource
Sociotechnical Safety Evaluation of Generative AI Systems
Original source
MIT AI Risk Repository
Open the public repository used for AI risk records and taxonomy fields.