Values Statement: We believe AI should cause no harm, but enhance the quality of human life, by proactively adopting our AI Governance framework.

Evidence-based Transparent For governance

AI Incidents

Critique of Government's AI Algorithm Performance: An 'F' Grade Analysis

This incident highlights concerns about governmental AI algorithm performance, emphasizing the importance of trustworthy AI governance in fo...

AI Bias Alert: UK Passport Photo Checker Shows Preference Against Dark-Skinned Women - Harm Prevention in AI

Unveiling an AI incident that highlights the importance of responsible AI governance: a UK passport photo checker demonstrates bias against...

Image Algorithm Misidentifies Jewish Baby Stroller: Highlighting Need for Trustworthy AI

This incident showcases the importance of responsible AI governance in preventing harm. A popular image algorithm misclassified a Jewish bab...

YouTube's Radicalization Trap: The Christchurch Shooter Incident and Responsible AI

Exploring the role of AI in the radicalization process, this article investigates the Christchurch shooter incident on YouTube. Understandin...

Stanford Vaccine Distribution: Prioritizing Medical Residents

A review of Stanford's initial vaccine distribution plan highlights the prioritization of medical residents among the first 5,000 doses allo...

Gender Bias Allegations Highlight the Need for Trustworthy AI in Fintech: Apple Card Incident

The Apple Card incident sheds light on the potential risks of biased AI in financial technology. This AI incident maps to the Govern functio...

Facebook Accused of Enabling Housing Discrimination: An Examination Under Responsible AI

This incident highlights the importance of trustworthy AI in preventing harm. The Department of Housing and Urban Development (HUD) has char...

Deliveroo Uses Discriminatory AI Algorithm: Court Ruling Highlights Need for Safe and Secure AI

A recent court ruling found Deliveroo's algorithm discriminated against certain riders, underscoring the importance of trustworthy AI. This...

Facial Analysis in Job Screening Services Suspended - A Step Towards Safe AI

Recent halt in the use of facial analysis by a job screening service signifies progress in responsible AI governance. This incident maps to...

Lawsuit Over AI-Based Teacher Evaluations in Houston Schools

An ongoing legal dispute challenges the use of AI in teacher evaluations, raising important questions about responsible AI governance. This...

Misinterpretation of Traffic Signals by Tesla's Autopilot: An AI Incident Highlighting the Importance of Responsible AI

This AI incident, involving Tesla's Autopilot system misreading red traffic lights as green, underscores the need for safe and secure AI. It...

Dangers of Inappropriate AI-Generated Content Targeted at Children - A Call for Responsible AI Governance

This disturbing incident involving inappropriate YouTube videos highlights the urgent need for responsible AI governance. These videos, gene...

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.