Incidente 114: Rekognition de Amazon comparó falsamente a miembros del Congreso con fotos policiales
Descripción: La ACLU demostró que la función de comparación de rostros de Rekognition identificó erróneamente a miembros del Congreso, y particularmente a miembros de color, como otras personas que habían sido arrestadas usando una base de datos de fotografías policiales creada a partir de fotos de arrestos disponibles públicamente.
Entidades
Ver todas las entidadesPresunto: un sistema de IA desarrollado e implementado por Amazon, perjudicó a Rekognition users y arrested people.
Clasificaciones de la Taxonomía CSETv1
Detalles de la TaxonomíaIncident Number
The number of the incident in the AI Incident Database.
114
Special Interest Intangible Harm
An assessment of whether a special interest intangible harm occurred. This assessment does not consider the context of the intangible harm, if an AI was involved, or if there is characterizable class or subgroup of harmed entities. It is also not assessing if an intangible harm occurred. It is only asking if a special interest intangible harm occurred.
yes
Notes (AI special interest intangible harm)
If for 5.5 you select unclear or leave it blank, please provide a brief description of why.
You can also add notes if you want to provide justification for a level.
The ACLU's test demonstrated Rekognition's disproportionate inaccuracy on the faces of people of color.
Date of Incident Year
The year in which the incident occurred. If there are multiple harms or occurrences of the incident, list the earliest. If a precise date is unavailable, but the available sources provide a basis for estimating the year, estimate. Otherwise, leave blank.
Enter in the format of YYYY
2018
Date of Incident Month
The month in which the incident occurred. If there are multiple harms or occurrences of the incident, list the earliest. If a precise date is unavailable, but the available sources provide a basis for estimating the month, estimate. Otherwise, leave blank.
Enter in the format of MM
07
Estimated Date
“Yes” if the data was estimated. “No” otherwise.
No
Risk Subdomain
A further 23 subdomains create an accessible and understandable classification of hazards and harms associated with AI
1.1. Unfair discrimination and misrepresentation
Risk Domain
The Domain Taxonomy of AI Risks classifies risks into seven AI risk domains: (1) Discrimination & toxicity, (2) Privacy & security, (3) Misinformation, (4) Malicious actors & misuse, (5) Human-computer interaction, (6) Socioeconomic & environmental harms, and (7) AI system safety, failures & limitations.
- Discrimination and Toxicity
Entity
Which, if any, entity is presented as the main cause of the risk
AI
Timing
The stage in the AI lifecycle at which the risk is presented as occurring
Post-deployment
Intent
Whether the risk is presented as occurring as an expected or unexpected outcome from pursuing a goal
Unintentional
Informes del Incidente
Cronología de Informes
La tecnología de vigilancia facial de Amazon es objeto de una creciente oposición en todo el país y, en la actualidad, hay 28 motivos más de preocupación. En una prueba que la ACLU realizó recientemente de la herramienta de reconocimiento f…
Variantes
Una "Variante" es un incidente de IA similar a un caso conocido—tiene los mismos causantes, daños y sistema de IA. En lugar de enumerarlo por separado, lo agrupamos bajo el primer incidente informado. A diferencia de otros incidentes, las variantes no necesitan haber sido informadas fuera de la AIID. Obtenga más información del trabajo de investigación.
¿Has visto algo similar?
Incidentes Similares
Did our AI mess up? Flag the unrelated incidents
Predictive Policing Biases of PredPol
· 17 informes
Northpointe Risk Models
· 15 informes
Incidentes Similares
Did our AI mess up? Flag the unrelated incidents
Predictive Policing Biases of PredPol
· 17 informes
Northpointe Risk Models
· 15 informes