Twitter’s Image Cropping Tool Allegedly Showed Gender and Racial Bias

September 18, 2020

Twitter's photo cropping algorithm, upon scrutiny by researchers, was found to show a preference towards white and female faces in images containing multiple faces. As a response, the company has discontinued its use on mobile platforms. This incident underscores the need for robust governance and guardrails for AI, as well as continuous measurement to prevent such harm. To join Project Cerebellum's efforts towards shaping safe and secure AI practices, visit JOIN US. Additionally, this incident aligns with our HISPI Project Cerebellum TAIM (Govern) approach.

Matched TAIM controls

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

MEASURE 2.10 — similarity 0.624, rank 1. TAIM detail and related incidents →
MAP 1.6 — similarity 0.620, rank 2. TAIM detail and related incidents →
GOVERN 3.1 — similarity 0.615, rank 3. TAIM detail and related incidents →

Alleged deployer: twitter
Alleged developer: twitter
Alleged harmed parties: twitter-users, twitter-non-white-users, twitter-non-male-users

Source

Data from the AI Incident Database (AIID). Cite this incident: https://incidentdatabase.ai/cite/103

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.