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

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

A recent study reveals racial, gender, and socioeconomic biases in chest X-ray classifiers. This underscores the need for trustworthy AI and...

Football Match AI Mishap: Ref's Bald Head Mistakenly Identified as Ball - Emphasizing Importance of Responsible AI Governance

This amusing incident highlights the need for safe and secure AI systems. This AI mistake, where an AI system failed to differentiate a refe...

AI Algorithm Bias Impact: Unfair Barrier to Kidney Transplants for Black Patients - Raising Awareness for Responsible AI

This AI incident highlights the potential harm that biased algorithms can cause in critical healthcare decisions, such as kidney transplants...

The Unseen AI Decision Making Algorithm in Higher Education: A Look at Its Impact

Explore the role of an algorithm in college admission decisions, shedding light on its impact on students. This AI incident maps to the Gove...

Clash of Expectations: The Reality vs Perception Gap in Robotics – An AI Incident Perspective

This AI incident, involving a robot in a public park, highlights the disconnect between our expectations and the current capabilities of rob...

Tracking Minors using Facial Recognition: An Examination of AI Incident for Criminal Suspects

This AI incident underscores the potential risks and ethical implications of facial recognition technology. As we strive towards trustworthy...

Lawsuit Filed Over Google Instant's Alleged Anti-Semitic Search Results: A Case for Responsible AI

This incident highlights the need for safe, secure, and trustworthy AI. The case against Google Instant's allegedly biased results underscor...

Misidentified by AI: The Case for Responsible Algorithm Design

This AI incident underscores the need for trustworthy algorithms. Join us to help prevent such errors, ensuring safe and secure AI. This AI...

Exploring Potential Biases in Pokémon Go: A Case Study on Redlining Communities of Color

This AI incident raises concerns about the potential for geospatial data bias in augmented reality apps such as Pokémon Go. It maps to the G...

Robot Employee Causes Unease Among Store Patrons: An Incident Illustrating the Need for Responsible AI Governance

This AI incident highlights the importance of trustworthy and safe AI. The misbehavior of the robot employee, which led to a decrease in cus...

Unauthorized Chatbots Shutdown in China: A Case for Safe and Secure AI Governance

An incident involving unauthorized chatbots in China raises concerns over responsible AI governance. These rogue bots, which questioned the...

Dangerous AI Reward Functions Gone Awry: A Case Study

Explore a real-world incident involving malfunctioning reward functions in an AI system, highlighting the importance of responsible AI gover...

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.