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

Examining the AI Bias in Organ Allocation Algorithms: A Focus on Kidney Transplants

Investigating an algorithm's discriminatory impact on black patients receiving kidney transplants, this incident underscores the importance...

Humorous AI Mishap: Football vs. Ref's Bald Head - A Lighthearted Reminder of the Need for Responsible AI

This AI mishap, where a football was incorrectly identified as a referee's bald head, showcases the importance of safe and secure AI. This i...

Uncovering Bias in Chest X-Ray Classifiers: A Call for Responsible AI

This AI incident underscores the importance of safe and secure AI, particularly in the medical field. Researchers discovered racial, gender,...

The Lekki Massacre: Investigating the Misinformation on Facebook - A Call for Responsible AI

In October 2020, a controversial incident occurred at Lekki Toll Gate in Lagos, Nigeria. However, many posts about this event were labeled '...

Unveiling the Controversy Behind Efficient Spam Filters: An Analysis

Exploring the often-overlooked ethical implications of spam filters, this article highlights the need for responsible AI governance in ensur...

Facebook Fails to Detect Misinformation on COVID-19 and Elections: A Reminder of the Need for Trustworthy AI

This incident highlights the challenges in detecting misinformation, emphasizing the importance of robust AI governance for safe and secure...

AI System Unintendedly Expresses Potential Harm: An Incident Highlighting the Importance of AI Governance

This AI incident underscores the necessity of robust governance for ensuring safe and secure AI operations. The system, intended to alleviat...

Critical Review: Assessing Algorithmic Decision-making in Public Sector - The Case of the Leaving Cert Exam

This AI incident, related to algorithmic decision-making in education, highlights the need for robust and trustworthy AI governance. It unde...

Bias in UK Passport Photo Checker: An Example of Untrustworthy AI

A recent incident involving the UK passport photo checker revealed a concerning bias against dark-skinned women. This AI incident maps to th...

Analyzing AI Incident: Jewish Baby Stroller Image Recognition Error

This AI incident sheds light on the need for trustworthy AI, highlighting a misclassification of a Jewish baby stroller image in an algorith...

YouTube's Radicalization Trap: The Christchurch Shooter Incident - Understanding Governance in Trustworthy AI

Explore the consequences of YouTube's algorithm on the Christchurch shooter incident, a stark example of potential harm from ungoverned AI....

Apology Issued for Potential Inequities in COVID-19 Vaccine Allocation: A Case Study on AI Governance

In this incident, Stanford University apologized for a coronavirus vaccine plan that prioritized some front-line doctors over others. This A...

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.