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

TAIM catalog Map Data source & citation

The Simplification of Deepfakes: A Growing Concern for Responsible AI Governance

Explore the alarming ease with which deepfakes can be created, their potential impacts on trustworthy AI, and the role of safe and secure AI...

Artificial Intelligence Generates Fake Speech Attributed to Former President Obama

An AI-generated speech, attributed to former U.S. President Barack Obama, was circulated online. The speech did not originate from the forme...

In-depth Analysis of the COMPAS Recidivism Algorithm: Promoting Responsible AI

This article delves into a comprehensive analysis of the much-debated COMPAS recidivism algorithm, highlighting its impact on fairness and t...

Examining AI Models for Fairness: A Crucial Step Towards Responsible AI

In the rapidly evolving landscape of Artificial Intelligence, ensuring fairness has emerged as a key concern. This article sheds light on th...

Are Bail Algorithms as Effective as Random Guesses?

A recent study published in the Proceedings of Machine Learning Research has questioned the accuracy of bail algorithms used by courts acros...

AI's Predictive Capabilities for Recidivism Fall Short of Human Performance

Exploring the limitations of AI in predicting recidivism, a study revealed its results were comparable to human judgement. The findings high...

AI's Limitations in Identifying Recidivism: Ineffective Predictions by Algorithms

A recent study has uncovered a surprising fact - algorithms used to predict repeat offenders are no more accurate than inexperienced humans....

Algorithm Performance in Predicting Recidivism Compared to Human Judgement

A recent study by the National Institute of Justice challenges the traditional notion that AI-driven algorithms are more accurate than human...

Fairness in AI: Assessing the Role of AI in Criminal Justice

Artificial Intelligence (AI) is increasingly being employed to aid in criminal justice, from convicting criminals to determining jail senten...

Addressing Racial Bias in Criminal Justice AI: A Responsible Approach

Exploring the ongoing debate around eliminating racial bias in criminal justice algorithms, this article highlights the need for responsible...

Mechanical Turkers Outperform COMPAS, a Notable Judicial AI Algorithm

In a recent study, human workers on Mechanical Turk outperformed the popular risk assessment tool, COMPAS, in predicting recidivism rates. T...

Questionable Accuracy of Court Software Predicting Criminal Risk Compared to Web Survey Participants

A recent study challenges the reliability of court software used for predicting criminal risk, suggesting it may be no more accurate than su...

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.