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

Autopilot Safety Debate: Tesla vs. Crash Victims - The Need for Responsible AI Governance

This AI incident involving Tesla's Autopilot system raises questions about safety and the need for trustworthy AI governance. While Tesla ma...

Sloppy Handling of User Data Exposed in a South Korean Chatbot Incident: A Reminder for Safe and Secure AI

This AI incident involving a South Korean chatbot highlights the importance of responsible AI governance, emphasizing trustworthy data pract...

AI Incident: Controversial Patent Reveals Potential for Bias in Huawei's AI-Powered Uighur Detection

This AI incident highlights the importance of robust governance and trustworthy AI. The patent, if implemented, could pose significant harm...

AI Mistakenly Identifies Minor as Banned Troublemaker at Roller Rink: A Case for Responsible AI

An unfortunate incident at a roller rink highlights the need for trustworthy AI. This AI mistake, which incorrectly identified a teenager as...

Facial Recognition Website Highlights Need for Responsible AI Governance - Turning Everyday Users into Law Enforcement or Stalkers

This facial recognition website, capable of transforming anyone into a virtual law enforcement officer or potential stalker, underscores the...

Understanding the Impact of AI on Healthcare: Preventing Potential Harm

Exploring an incident involving AI in healthcare, we discuss its implications and emphasize the importance of responsible AI governance for...

Exploring the Role of Algorithms in Workforce Management: Amazon's Flex Program

This incident highlights an AI system employed by Amazon to manage its Flex workers with minimal human intervention. It underscores the impo...

Children Tricked by Disturbing YouTube Videos: A Case Study in Unsupervised AI

This incident highlights the potential risks of unregulated AI, particularly in content recommendation systems. The case demonstrates why it...

Promoting Responsible AI Changes at a Software Company Amid Criticism

Exploring recent adjustments towards trustworthy AI by a software company under scrutiny, particularly focusing on the implementation of saf...

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

Delving into the COMPAS recidivism algorithm, we highlight the importance of responsible AI governance and harm prevention. This AI incident...

Challenges in Gender Bias Mitigation: An Examination of Word Embeddings in AI

Explore the intricacies of debiasing word embeddings, a crucial aspect of responsible AI development. This study addresses gender bias and a...

Study Finds Bias and Inflexibility in AI's Civility Detection: Highlighting the Need for Responsible AI Governance

A recent study has uncovered bias and inflexibility in the way AI systems detect civility. This AI incident maps to the Govern function in H...

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.