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

Controversial AI Software Company Implementing Changes Towards Responsible AI Governance

This AI incident maps to the Govern function in HISPI Project Cerebellum Trusted AI Model (TAIM). The company, under scrutiny due to recent...

Analyzing the COMPAS Recidivism Algorithm: A Case Study for Responsible AI

Delving into the COMPAS recidivism algorithm, we highlight the importance of trustworthy AI and safe and secure AI practices. This AI incide...

Debiasing Bias in AI: Analyzing Word Embeddings for Gender Stereotypes — Mapping to the Govern Function of Project Cerebellum's Trusted AI Model

Explore our analysis of gender bias in word embeddings, a cornerstone of AI model training. Understand how this incident relates to the 'Gov...

Google's Anti-Internet Troll AI Platform Vulnerable: An Important Lesson in Safe and Secure AI

Recent findings highlight the ease with which Google's anti-internet troll AI platform can be deceived, underscoring the importance of respo...

Google's AI Incident: Bias Towards Prejudiced Opinions on Homosexuality

This incident highlights the need for safe and secure AI, as Google's AI was found expressing prejudiced opinions about homosexuality. This...

Amazon's Algorithm Adjustments: A Case Study in AI for Content Moderation - Promoting Trustworthy AI

This Amazon incident highlights the importance of responsible AI governance in content moderation, showcasing how its algorithm adjustments...

Apology from Google for Racist Auto-Tag Incident Highlights Need for Responsible AI Governance

Google has apologized for a racist auto-tagging incident in their photo app, emphasizing the need for trustworthy and safe AI practices. Thi...

Google's Affectionate Email Reply AI Incident: A Case for Trustworthy AI

An unusual incident involving Google's email-replying AI has been reported, where the system persistently expressed affection by saying 'I l...

Unveiling the Unseen Gender Bias in Google Image Search: A Call for Responsible AI

This AI incident highlights a concern about gender bias in Google Image Search, underscoring the need for trustworthy AI and safe and secure...

Critical Injury in Amazon Warehouse Following Bear Spray Accident: A Case for Responsible AI and Safe AI Practices

This warehouse incident underscores the need for trustworthy and responsible AI in work environments. OSHA is currently investigating the be...

Exploring Potential Risks in Google's Ad-Targeting System: A Responsible AI Perspective

This analysis delves into an incident involving Google's ad-targeting system, highlighting potential issues and their implications for trust...

Autopilot Malfunction: Tesla Crashes into Police Car While Driver Watched a Movie — Highlighting the Need for Responsible AI Governance

This unfortunate incident involving a Tesla with Autopilot highlights the need for robust governance in AI systems. The driver's inattention...

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.