Incident 167: Researchers' Homosexual-Men Detection Model Denounced as a Threat to LGBTQ People’s Safety and Privacy
Description: Researchers at Stanford Graduate School of Business developed a model that determined, on a binary scale, whether someone was homosexual using only his facial image, which advocacy groups such as GLAAD and the Human Rights Campaign denounced as flawed science and threatening to LGBTQ folks.
Entities
View all entitiesAlleged: Michal Kosinski and Yilun Wang developed and deployed an AI system, which harmed LGBTQ people , LGBTQ people of color and non-American LGBTQ people.
Incident Stats
Incident ID
167
Report Count
1
Incident Date
2017-09-07
Editors
Sean McGregor, Khoa Lam
GMF Taxonomy Classifications
Taxonomy DetailsKnown AI Goal
An AI Goal which is almost certainly pursued by the AI system referenced in the incident.
Behavioral Modeling
Known AI Technology
An AI Technology which is almost certainly a part of the implementation of the AI system referenced in the incident.
Neural Network
Potential AI Technology
An AI Method / Technology which probably is a part of the implementation of the AI system referenced in the incident.
Siamese Network, Convolutional Neural Network, Diverse Data
Known AI Technical Failure
An AI Technical Failure which almost certainly contributes to the AI system failure referenced in the incident.
Limited Dataset, Dataset Imbalance, Generalization Failure
Potential AI Technical Failure
An AI Technical Failure which probably contributes to the AI system failure referenced in the incident.
Incomplete Data Attribute Capture, Overfitting, Lack of Explainability
Incident Reports
Reports Timeline
nytimes.com · 2017
- View the original report at its source
- View the report at the Internet Archive
Michal Kosinski felt he had good reason to teach a machine to detect sexual orientation.
An Israeli start-up had started hawking a service that predicted terrorist proclivities based on facial analysis. Chinese companies were developing fac…
Variants
A "variant" is an incident that shares the same causative factors, produces similar harms, and involves the same intelligent systems as a known AI incident. Rather than index variants as entirely separate incidents, we list variations of incidents under the first similar incident submitted to the database. Unlike other submission types to the incident database, variants are not required to have reporting in evidence external to the Incident Database. Learn more from the research paper.
Similar Incidents
Did our AI mess up? Flag the unrelated incidents
Similar Incidents
Did our AI mess up? Flag the unrelated incidents