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

Understanding an AI Duplicate Incident: Implications for Trustworthy AI

Apr 17, 2021

Dive into the technical aspects of a recent AI duplicate incident, and learn how this maps to the Govern function in HISPI Project Cerebellu...

Fatal Tesla Accident Highlights Importance of Safe and Secure AI - This AI Incident Maps to the Govern Function in HISPI Project Cerebellum Trusted AI Model (TAIM)

Apr 17, 2021

A 2019 Tesla Model S, operating under Adaptive Cruise Control (ACC), reportedly crashed into a tree near The Woodlands in Spring, Texas, res...

Inaccurate Auto-Insurance Photo Estimates: A Threat to Safe and Secure AI?

Apr 13, 2021

Repair shop owners and industry groups claim that auto-insurance companies' photo-based repair price estimation can lead to inaccurate estim...

GPT-3 Powered Game Abused: Generating Sexually Explicit Content - Harm Prevention Needed

Apr 1, 2021

Recent reports indicate that Latitude's AI Dungeon, powered by GPT-3, was manipulated by certain players to generate sexually explicit conte...

Uber Eats Driver's Lawsuit Highlights Importance of Safe and Secure AI: Excessive Photo Checks and False Mismatches

Apr 1, 2021

A former Uber Eats driver filed a lawsuit, claiming the company unjustly dismissed him due to frequent false matches of his verification sel...

Tesla Autopilot Incident: Unattended Police Car Collision Highlights Need for Responsible AI Governance

Mar 17, 2021

A Tesla Model Y utilizing Autopilot crashed into a parked Michigan State Police car, equipped with emergency lights, in Eaton County, Michig...

French Welfare Office's Automated System Issues Unwarranted 'Robo-Debt': A Case of Untrustworthy AI

Mar 17, 2021

A woman receiving welfare benefits in France was incorrectly notified of a €542 debt due to an automated software system within the welfare...

Alleged Misuse of Internal Software at Dartmouth's Geisel School of Medicine: A Case for Responsible AI in Exam Systems

Mar 15, 2021

An internal software system at Dartmouth's Geisel School of Medicine has been accused of incorrectly flagging students for cheating during r...

AI Governance: Amazon's Alleged Speed Prioritization System in Delivery Drivers' Crash Raises Concerns

Mar 15, 2021

A lawsuit points to Amazon, citing liability for a crash involving its delivery driver. The suit alleges that the company's AI-powered drive...

Forced Deployment of AI Cameras on Amazon Delivery Drivers: A Concern for Trustworthy AI

Mar 2, 2021

AI governance is crucial in the deployment of AI-powered cameras on Amazon delivery drivers, as their location, movement, and biometric data...

Year-long AI Surveillance Pilot in Two South Australian Aged Care Facilities Reportedly Overwhelmed Staff with False Positives

Mar 1, 2021

Between March 2021 and March 2022, an AI-enabled video and audio monitoring system was trialed in two South Australian aged care facilities....

Ineffective Automated Moderation: Facebook's Hate Speech, Violence, and Incitement Issue - A Call for Responsible AI

Mar 1, 2021

Internal reports reveal that Facebook's automated moderation tools are significantly outperformed by human moderators. They account for a me...

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.