The Case Study of Microsoft's Tay: A Lesson in Responsible AI
Exploring the infamous AI incident involving Tay, Microsoft's chatbot that learned and perpetuated offensive language. This AI incident maps...
Read moreValues 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
Data source & citationAI-Powered Mall Security Robot Causes Accident: A Call for Responsible AI Governance
An unfortunate incident involving a toddler and an AI-powered mall security robot in Silicon Valley highlights the need for robust AI govern...
Read moreTesla Autopilot Incident Involving Joshua Brown Highlights Need for Safe and Secure AI
The tragic accident involving Joshua Brown, who died while operating his Tesla in autopilot mode, underscores the importance of trustworthy...
Read moreExamining Google's Image Search Bias Incident: A Case Study on Responsible AI
This incident highlights the importance of trustworthy AI, where an analysis of image search results for black teenagers demonstrated potent...
Read moreUnderstanding and Addressing Machine Bias: A Key Aspect of Trustworthy AI
Exploring the impact and causes of machine bias, this article emphasizes the importance of responsible AI governance in ensuring fairness an...
Read moreMinor's Exposure to Inappropriate Content Highlights Need for Responsible AI - AI Incident Maps to Govern Function in Project Cerebellum Trusted AI Model (TAIM)
A distressing incident occurred when a child asked a popular digital assistant for a song, instead receiving pornographic content. This unde...
Read moreIncident Analysis: Amazon's AI Mishap in Mobile Case Production - Emphasizing Responsible AI
This AI incident involving Amazon's cell phone case production serves as a reminder of the importance of safe and secure AI. The mishap high...
Read moreExamining the Impact of Automated Robodebt on Governance: A Case Study in AI Incident Database
This AI incident, marking the adoption of robodebt by Human Services, highlights the potential challenges that arise when humans are removed...
Read moreUnderstanding the Yandex Chatbot Incident: A Step Towards Responsible AI
Explore the recent incident involving the Yandex chatbot, a crucial learning experience in our journey towards trustworthy AI. This AI incid...
Read moreGoogle Translate's Error: Gender Bias in AI-Generated Languages
A recent incident involving Google Translate revealed a gender bias, generating inappropriate translations for 'female historians' and 'male...
Read moreApology Issued for Controversial Skin-Toning Filter in FaceApp: An Example of AI Governance
FaceApp, a popular photo editing app, has issued an apology for a 'racist' filter that lightens users' skin tones. This incident underscores...
Read moreEdit Wars on Improved Wikipedia by AI Bots Highlight the Need for Responsible AI Governance
This incident involving AI bots on Wikipedia demonstrates the importance of safe, secure and trustworthy AI governance. The squabbling edit...
Read moreData 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.