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

Accidental YouTube Algorithm Block: A Case Study in AI Governance and Trustworthy AI

An examination of an unintended algorithmic blockade on the 'black v white' CHESS strategy video, shedding light on the importance of respon...

Experience with Tesla's 'Full Self-Driving': Understanding AI Safety Issues

Exploring a real-world scenario involving Tesla's Full Self-Driving feature, this article highlights the importance of safe and secure AI. T...

Exploring the Ethics of AI: Can Artificial Intelligence Understand Morality?

This AI incident maps to the Govern function in HISPI Project Cerebellum Trusted AI Model (TAIM). In this article, we delve into the complex...

Bank Fraud: Voice Cloning Leads to $35 Million Loss - A Reminder of the Need for Responsible AI Governance

In an alarming incident, fraudsters successfully cloned a company director's voice to execute a $35 million bank heist. This underscores the...

False Claims in Website Accessibility Overlays: Understanding the Impact on Responsible AI

Exploring a case of misleading claims in website accessibility overlays, this AI incident highlights the need for safe and secure AI. This A...

Zillow Exits Home Buying Business: Impact on AI Governance and Responsible AI Development

Zillow's decision to exit its home buying business and reduce staff by 25% raises questions about the role of AI in such businesses. This AI...

YouTube Videos Tricking Children: A Case Study in AI Incident Governance

This incident highlights the importance of responsible AI governance, as children were exposed to potentially harmful content on YouTube. Th...

Reevaluating AI Ethics: Changes Made by the Contentious Software Company in Response to Criticism

This AI incident sheds light on the company's efforts towards responsible AI governance, demonstrating their commitment to trustworthy and s...

Analyzing the COMPAS Recidivism Algorithm: A Case Study on Trustworthy AI

In this article, we delve into the analysis of the controversial COMPAS recidivism algorithm. This investigation underscores the importance...

Debiasing Word Embeddings: A Step Towards Gender-Fair AI - Improving Trustworthy AI through Project Cerebellum

Explore the challenges of debiasing word embeddings in AI and how it contributes to gender fairness. This AI incident maps to the Govern fun...

Study Reveals Bias and Inflexibility in AI's Civility Detection: A Call for Safe and Trustworthy AI

A recent study highlights the shortcomings of AI in civility detection, raising concerns about its bias and inflexibility. This AI incident...

Exploring Biases in Google's AI Regarding Homosexuality: A Case for Responsible AI

This incident showcases the potential biases in AI systems, highlighting the need for trustworthy AI and safe & secure AI governance. By rep...

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.