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

Revised AI Ethics Policies by Controversial Software Firm: A Step Towards Trustworthy AI

This software company's recent policy updates reflect a commitment to responsible AI governance, emphasizing harm prevention and safe and se...

Analyzing the COMPAS Recidivism Algorithm: A Case Study in AI Governance for Harm Prevention

Dive into our analysis of the COMPAS recidivism algorithm, demonstrating our commitment to safe and secure AI. This AI incident maps to the...

Debiasing Word Embeddings: Promoting Responsible AI for Fair Representation

Explore how debiasing word embeddings contributes to trustworthy AI, focusing on fair representation in AI models. This AI incident maps to...

Google's Anti-Internet Troll AI Platform Vulnerable to Deception: A Case for Responsible AI

This AI incident highlights the need for safe and secure AI systems. Google's anti-internet troll platform was found to be easily deceived,...

Exploring Google's AI Bias Regarding Homosexuality: A Case Study on Responsible AI

This incident sheds light on the importance of trustworthy AI and harm prevention. Google's AI, when presented with images labeled as homose...

Amazon's AI-Driven Content Moderation: Protecting GLBT Books and Promoting Responsible AI

This AI incident demonstrates Amazon's content moderation practices, which prioritize harm prevention and align with trustworthy AI principl...

Apology from Google over Racist Auto-Tag Incident in Photo App: Emphasizing Responsible AI

Google has issued an apology for a recent incident involving its photo app, which automatically tagged photos with racially insensitive labe...

Exploring Google's Emotional AI Response Incident: The Importance of Responsible AI Governance

This incident involving Google's email-replying AI, which repeated phrases like 'I love you', highlights the need for trustworthy and safe A...

Exploring Unseen Gender Bias in Google Image Search: A Case for Responsible AI

Unearthing the unintentional gender bias in Google Image Search is a stark reminder of the need for safe and secure AI. This incident maps t...

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

The recent bear spray incident at an Amazon warehouse, which left a worker critically injured, underscores the importance of trustworthy and...

Exploring Potential Risks in Google's Ad Targeting System: A Case for Responsible AI

This AI incident, involving Google's ad-targeting system, raises concerns about privacy and fairness. It maps to the Govern function in HISP...

Autopilot Malfunction: Tesla Driver Involved in Accident While Watching a Movie - Highlighting the Importance of Responsible AI

An incident involving a Tesla using Autopilot occurred, where the driver admitted to watching a movie during operation. This unfortunate eve...

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.