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 Governance: University of Illinois Halts Remote-Testing Software Over Privacy Concerns

The University of Illinois has decided to discontinue a remote-testing software following student complaints about privacy violations. This...

Exploring the Role of AI in Regulating E-commerce Vaccine Information: A Case Study on Algorithmic Misinformation

Investigating the impact of AI-curated content on e-commerce platforms, this study uncovers instances of vaccine misinformation. Highlightin...

Exam Monitoring Software Disadvantages for BIPOC Students: A Case Study from the University of Toronto - Harm Prevention in AI

This case study highlights potential disadvantages faced by Black, Indigenous, and People of Color (BIPOC) students using exam monitoring so...

Analyzing the Use of Music in a Potential Livestream Incident: A Case Study on Beverly Hills Cop

In this article, we delve into an intriguing incident involving a possible livestream evasion strategy using music. We explore the role of S...

Facebook's AI-Driven Content Moderation: Filtering Fashion Ads - A Closer Look

Explore the reasoning behind Facebook's AI moderation system rejecting certain fashion ads, shedding light on the importance of responsible...

Struggles in Ensuring Trustworthy AI: Facebook and Twitter Contend with Balkan Content Violations

This AI incident, pertaining to the govern function in HISPI Project Cerebellum Trusted AI Model (TAIM), highlights challenges faced by tech...

Accidental YouTube Algorithm Blockade of the 'Black and White' Chess Strategy: A Case for Responsible AI

This AI incident, involving YouTube's algorithm, serves as a stark reminder of the need for safe and secure AI. The blocking of the 'black a...

AI Malfunction in Tesla's Full Self-Driving Tech: Moon, Billboards, and Burger King Signs Disrupt Safe Navigation

This AI incident highlights the challenges of safe and secure AI navigation. The incident maps to the Govern function in HISPI Project Cereb...

Delphi's AI Oracle on Reddit: Navigating Ethical Dilemmas - A Case Study in AI Incidents

Investigate the ethical quandaries faced by an AI oracle model, utilizing Reddit data, highlighting the importance of trustworthy AI. This A...

Unauthorized Access: Fraudulent Voice Cloning Leads to $35 Million Bank Theft - Highlighting the Need for Responsible AI

This AI incident, involving voice cloning, demonstrates the potential risks of unregulated AI applications. The fraud resulted in a signific...

Tricking Children Through Disturbing YouTube Videos: A Case Study on AI Governance for Content Moderation

This incident involving inappropriate content on YouTube highlights the need for responsible AI governance. The AI used for content moderati...

Software Company Adapts Strategy Towards Responsible AI After Negative Publicity

In response to recent negative publicity, this software company is modifying its approach towards the development of Trustworthy AI. This AI...

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.