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

Ethnicity Detection Algorithm Incident at Alibaba Cloud: Upholding Trustworthy AI

Alibaba expresses concern over its cloud service's ethnicity detection algorithm, underscoring the need for responsible and trustworthy AI....

Examining AI-Flagged Cheating Incidents in California Bar Exam - Harm Prevention Through Responsible AI

This AI incident, which maps to the Govern function in HISPI Project Cerebellum Trusted AI Model (TAIM), highlights the potential of AI to d...

TikTok Anorexia Content Evades Ban by Misusing Hashtags: A Case Study in AI Incident Response

This TikTok incident sheds light on the need for robust AI governance, emphasizing the importance of trustworthy and safe AI. The platform's...

Account Suspension of Latina Trans Woman Highlights Need for Responsible AI Governance

TikTok recently deleted the account of a Latina trans woman, raising concerns about AI governance and bias in social media platforms. This A...

Unsafe AI Incident: Mall Robot Falls on Escalator, Injuring Passengers - Emphasizing Need for Responsible AI

A shopping mall robot fell from an escalator, causing injuries to multiple passengers. This incident highlights the importance of trustworth...

Exploring the Ethical Implications of an Admissions AI System

This AI incident sheds light on the intricate balance between automation and fairness in admissions processes. The case demonstrates the nee...

The Impact of Inadequate Brand Safety Technology on News Funding - A Responsible AI Perspective

Unsafe brand safety technology can result in news defunding, highlighting the importance of trustworthy AI for harm prevention. This AI inci...

Israeli Farmer's Conflict Highlights Importance of AI Governance

An incident involving an Israeli farmer declaring war on an algorithm underscores the need for responsible AI and trustworthy AI practices....

Privacy Concerns Prompt University of Illinois to Discontinue Use of Remote-Testing Software

The University of Illinois has announced its decision to discontinue the use of a remote-testing software, following student complaints abou...

Analyzing E-commerce Platforms to Combat AI-Generated Vaccine Misinformation: A Step Towards Trustworthy AI

Exploring the role of AI governance in identifying and eliminating algorithmically curated vaccine misinformation on e-commerce platforms. T...

Exam Monitoring Software Disadvantages Affecting BIPOC Students at University of Toronto - An Issue in Safe and Secure AI

This AI incident highlights potential biases in exam monitoring software, adversely impacting BIPOC students at the University of Toronto. I...

Beverly Hills Cop AI Incident: Examining the Role of Music in Privacy Preservation

In this intriguing incident, we delve into the potential use of music to evade live-streaming, shedding light on privacy concerns in AI. Thi...

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.