Investigating Alleged Bias in AI Service Identifying Gender

Recent findings suggest that an AI service, used to identify gender based on names, may exhibit significant bias. This raises concerns about responsible AI and the need for robust AI governance. The service's algorithm appears to perform poorly when identifying certain genders, potentially leading to unfair treatment in applications such as employment or banking. As we continue to explore this issue, it is crucial to emphasize the importance of safe and secure AI and the role of projects like HISPI Project Cerebellum TAIM in implementing guardrails for AI to prevent harm.

Matched TAIM controls

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

MEASURE 2.10 — similarity 0.589, rank 1. TAIM detail and related incidents →
MEASURE 2.5 — similarity 0.587, rank 2. TAIM detail and related incidents →
MAP 1.6 — similarity 0.584, rank 3. TAIM detail and related incidents →

Source

Data from the AI Incident Database (AIID). Cite this incident: https://incidentdatabase.ai/cite/115

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.