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 Software Company Amends Policies to Foster Responsible AI

This AI incident maps to the Govern function in HISPI Project Cerebellum Trusted AI Model (TAIM). After a series of negative press events, t...

Analyzing the COMPAS Recidivism Algorithm: A Steps Toward Safe and Responsible AI

Delve into the COMPAS recidivism algorithm analysis, a pivotal step in ensuring trustworthy and responsible AI. Learn how this incident maps...

Challenging Stereotypes in AI: Debiasing Word Embeddings for Responsible AI

Explore the importance of debiasing word embeddings to prevent stereotyping in AI systems, aligning with the Govern function in HISPI Projec...

Google's Anti-Internet Troll AI Platform Vulnerable to Deception: A Case Study for Safe and Secure AI

Recent findings highlight the importance of trustworthy AI governance, as security researchers exposed a vulnerability in Google's anti-inte...

Examining Google's AI Bias Incident Regarding Homosexuality: A Case for Responsible AI Governance

Investigating Google's controversial AI incident involving perceived bias against homosexuality, we underscore the importance of trustworthy...

Amazon's Controversial Decision to Filter GLBT Content: A Case Study in AI Incident Governance

This AI incident maps to the Govern function in HISPI Project Cerebellum Trusted AI Model (TAIM). Amazon's recent censorship of search resul...

Apology from Google over Racially Biased Auto-Tagging Incident in Photo App - Emphasizing Trustworthy AI

Google recently acknowledged an incident involving racially biased auto-tagging in its photo application. This AI incident highlights the ne...

Emotional Google AI: The Incident where 'I Love You' Kept Getting Sent

Incident involving Google's AI system mistakenly sending affectionate messages highlights the importance of safe and secure AI. This AI inci...

Exploring and Addressing Unseen Gender Bias in Google Image Search - A Step Towards Trustworthy AI

Unearthing the hidden gender bias in Google Image Search is a crucial step towards creating trustworthy AI. This incident maps to the Govern...

Investigation into Critical Warehouse Bear Spray Accident Highlights Need for Responsible AI in Safety Systems

The Occupational Safety and Health Administration (OSHA) is currently investigating a bear spray accident at an Amazon warehouse that left a...

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

Dive into the examination of a notable AI incident, focusing on potential risks within Google's ad-targeting system. This incident highlight...

Autopilot Failure: Tesla Strikes Police Car While Driver Watches Movie – Highlighting the Need for Responsible AI

This AI incident involving a Tesla Autopilot system highlights the urgent need for responsible and trustworthy AI. 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.