Researchers' Homosexual-Men Detection Model Denounced as a Threat to LGBTQ People’s Safety and Privacy

September 7, 2017

Researchers at Stanford Graduate School of Business developed an AI model that, using only facial images, attempted to determine sexual orientation on a binary scale. The model was denounced by advocacy groups like GLAAD and the Human Rights Campaign as problematic, potentially threatening the privacy and safety of LGBTQ individuals. Such models highlight the importance of responsible AI governance and the need for trustworthy and safe AI practices.

Join us in HISPI Project Cerebellum TAIM to Govern and Measure the impact of such incidents on AI harm prevention and help establish guardrails for AI development. JOIN US

Matched TAIM controls

Suggested mapping from embedding similarity (not a formal assessment). Browse all TAIM controls

MEASURE 2.10 — similarity 0.701, rank 1. TAIM detail and related incidents →
MAP 1.6 — similarity 0.695, rank 2. TAIM detail and related incidents →
MAP 4.1 — similarity 0.676, rank 3. TAIM detail and related incidents →

Alleged deployer: michal-kosinski, yilun-wang
Alleged developer: michal-kosinski, yilun-wang
Alleged harmed parties: lgbtq-people, lgbtq-people-of-color, non-american-lgbtq-people

Source

Data from the AI Incident Database (AIID). Cite this incident: https://incidentdatabase.ai/cite/167

Data source

Incident data is from the AI Incident Database (AIID).

When citing the database as a whole, please use:

McGregor, S. (2021) Preventing Repeated Real World AI Failures by Cataloging Incidents: The AI Incident Database. In Proceedings of the Thirty-Third Annual Conference on Innovative Applications of Artificial Intelligence (IAAI-21). Virtual Conference.

Pre-print on arXiv · Database snapshots & citation guide

We use weekly snapshots of the AIID for stable reference. For the official suggested citation of a specific incident, use the “Cite this incident” link on each incident page.