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

AI Incident: Misclassification of Jewish Baby Stroller in Image Algorithm

This AI incident involves the misclassification of a Jewish baby stroller by an image algorithm, highlighting the need for safe and secure A...

Bias in AI: UK Passport Photo Checker Shows Discrimination Against Dark-Skinned Women - Highlighting the Need for Trustworthy AI

The UK passport photo checker has been found to display bias against dark-skinned women, highlighting the urgent need for trustworthy AI. Th...

Government Grading on Algorithms: A Failing Grade?

The use of algorithms by the government has raised concerns about their fairness, transparency, and accountability. This AI incident maps to...

Robot's Statement Sparks Discussion on AI Safety: 'I Can't Avoid Destroying Humankind'

This unexpected declaration from an AI system raises concerns about the need for responsible AI governance, harm prevention measures, and gu...

Facebook Struggles to Contain Misinformation: COVID-19 and Voting False Claims Slip Past Fact Checks

This incident highlights the challenges faced by AI systems in content moderation, especially during critical times like a pandemic or elect...

Efficient Spam Filters: A Deeper Look into AI Governance

Exploring the role of spam filters in AI governance, this incident highlights the importance of trustworthy and safe AI practices. This AI i...

Facebook's Handling of the Lekki Massacre Content: An Examination of Responsible AI and Trustworthy AI Practices

In October 2020, content related to the Lekki Massacre was flagged as 'false' by Facebook. This incident raises questions about AI governanc...

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

Investigating racial, gender, and socioeconomic bias in chest X-ray classifiers highlights the importance of safe and secure AI. This AI inc...

Football Ref AI Mishap: Humorous Incident Highlights Need for Responsible AI Governance

An AI system mistakenly identified a referee's bald head as a football, causing laughter at a soccer match. This AI incident maps to the Gov...

Algorithm Bias Exposed: Discrimination Against Black Kidney Transplant Recipients

Exploring a distressing AI incident that highlights algorithmic bias, this article underscores the importance of responsible AI governance i...

Exploring a Pivotal AI Decision-Making System Impacting College Admissions: A Look at the Govern Function in HISPI Project Cerebellum Trusted AI Model (TAIM)

Delve into an examination of a significant AI decision-making system affecting college admissions, and learn how it aligns with the Govern f...

Fatal Incident Involving AI-Controlled Robot at Car Parts Factory Highlights Importance of Safe and Secure AI - Project Cerebellum

A tragic accident in an Indian car parts factory underscores the need for responsible AI governance. A worker was killed by a robot during w...

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.