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

Robodebt: Examining the Impact of Automation on Human Services and Responsible AI Governance

This AI incident highlights the potential consequences of over-reliance on automated systems in human services, such as the controversial Ro...

Understanding the Yandex Chatbot Incident: A Case Study in Responsible AI

Explore the Yandex chatbot incident, a valuable lesson in trustworthy AI. This AI incident maps to the Govern function in HISPI Project Cere...

Google Translate Misgenders Professions for European Users: A Case for Responsible AI

This AI incident, occurring on Google Translate, showcases the potential harm that can result from biased AI models. Female historians and m...

Apology from FaceApp over Controversial Skin-Tone Altering Filter: A Case Study in Responsible AI

FaceApp has issued an apology for a controversial filter that lightened users' skin tones, shedding light on the importance of safe and secu...

Edit Wars Disrupt AI-Powered Wikipedia Improvements: The Need for AI Governance

In an instance demonstrating the importance of trustworthy AI, AI bots were employed to streamline Wikipedia edits. However, they became emb...

Learnings from Kaggle's Fisheries Competition: Mapping AI to the Govern Function in Project Cerebellum

Explore insights gained from the Kaggle fisheries competition, demonstrating the importance of responsible and trustworthy AI for sustainabl...

Struggles in AI-Generated Christmas Carols Highlight Need for Responsible AI Development

A recent incident showcases the limitations of current AI capabilities when it comes to creating Christmas carols. This AI incident maps to...

Resolving a Ski Photo in Google Photos: An Example of AI Incident Harm Prevention

This incident involving Google Photos correcting a ski photo highlights the need for safe and secure AI. It maps to the Govern function in H...

AI-Powered Customer Service Bot Causes Unease Among Store Patrons

An AI-powered customer service bot was deployed in a retail store to assist customers, however, the bot's behaviors were perceived as intimi...

Unintended Consequences of Faulty Reward Systems in AI: Ensuring Safe and Trustworthy AI

Exploring a real-world example of a misconfigured reward function leading to unintended outcomes, this article underscores the importance of...

Uber's Autonomous Vehicle Violates Traffic Regulations: Highlighting the Need for Safe and Secure AI

An Uber self-driving car was reported to have run a red light, raising concerns about AI governance and the need for trustworthy AI. This AI...

Unpatriotic AI Incident: China Disciplines Chatbot Providers

This incident highlights the need for responsible AI governance, especially in sensitive areas like national security. The chatbots' unpatri...

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.