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

Deepfake of President Obama's Speech: An Unwanted Demonstration of AI Power

A deepfake video of former President Barack Obama delivering a speech has surfaced, highlighting the potential misuse and need for safeguard...

AI-Powered Bias in Criminal Justice Systems: An Unfair Sentence

An examination of a computer program's role in perpetuating racial bias within the criminal justice system, highlighting the importance of r...

Analyzing the Unchecked Spread of Violent Content in AI: A Case Study on MIT's AI Incident

This AI incident involving MIT scientists sheds light on a dangerous trend: feeding AI harmful content from Reddit, which can lead to unpred...

Exploring the AI Governance in the National Residency Matching Program - A Case Study for Safe and Secure AI

Dive into an analysis of the National Residency Matching Program, showcasing its role as a robust labor market model. This case study highli...

Exploring Unintended Bias: A Case Study on Racist AI

Delve into the unforeseen consequences of AI development: this AI incident serves as a cautionary tale, highlighting the importance of respo...

Understanding the Electric Elves Incident: Ensuring Responsible AI Operations

Explore the Electric Elves incident, a significant reminder of the importance of safe and secure AI operations. This AI incident maps to the...

Robotic Surgery Incidents Highlight Importance of Safe and Secure AI: 144 Deaths Since 2000

This AI incident in robotic surgery underscores the need for trustworthy AI. With a reported 144 deaths since 2000, it's crucial to establis...

Google Urged to Alter Autocomplete Function in Japan for Safe and Secure AI: A Case Study on Responsible AI Governance

This AI incident involving Google's autocomplete function highlights the importance of responsible AI governance. The case study maps to the...

Faulty Feature Halts Sale of Google's Nest Smart Smoke Alarm - Highlighting the Importance of Safe and Secure AI

Google's Nest has temporarily halted sales of its smart smoke alarm due to a faulty feature. This incident underscores the need for trustwor...

Investigating Alleged Bias in LinkedIn's Recommendation System: A Step Towards Responsible AI

Recent claims of gender bias in LinkedIn's recommendation system have raised concerns about fairness and transparency. This incident maps to...

Racially Biased AI Incident: New Zealand Passport Robot Mistakes Applicant of Asian Descent

A disturbing AI incident has been reported in New Zealand, where an applicant of Asian descent was instructed by a passport robot to 'open t...

Understanding Potential Consequences of AI System Failures: A Case Study

Explore the repercussions when AI systems malfunction, shedding light on the importance of safe and secure AI. This AI incident maps to the...

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.