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

TAIM catalog Map Data source & citation Failure Files™

Kenyan Data Annotators Allegedly Exposed to Graphic Content for OpenAI's AI

Nov 1, 2021

Kenyan contractors working for Sama AI allegedly faced excessive workloads annotating potentially disturbing content for OpenAI's generative...

California Regulator Suspended Pony.ai's Driverless Testing Permit Following a Non-Fatal Collision

Oct 28, 2021

The California Department of Motor Vehicles (DMV) suspended Pony.ai's driverless testing permit following a collision in San Francisco. The...

Research Prototype AI, Delphi, Reportedly Gave Racially Biased Answers on Ethics

Oct 22, 2021

The Delphi research model, a publicly accessible language-based AI trained via Reddit threads, reportedly demonstrated racial bias in its re...

Student-Developed Facial Recognition App Raised Ethical Concerns

Oct 21, 2021

TheFaceTag app, a social networking application developed by a student at Harvard University, has sparked discussions regarding facial recog...

Traffic Camera Misread Text on Pedestrian's Shirt as License Plate, Causing UK Officials to Issue Fine to an Unrelated Person

Oct 18, 2021

A local resident of Bath, UK, received an unwarranted fine due to a traffic camera misinterpreting text on her shirt as a license plate numb...

McDonald's AI Drive-Thru Allegedly Collected Biometric Customer Data without Consent, Violating BIPA

Oct 15, 2021

Lawsuit claims McDonald's chatbot in its Chicago AI drive-thru collected and processed voice data without user consent, potentially violatin...

Social Media Algorithms Amplified Disinformation Campaign in Honduras Election

Oct 6, 2021

In October 2021, a network of over 317 fake Twitter accounts used AI-driven algorithms to amplify disinformation about the Honduran presiden...

Facebook Internally Reported Failure of Ranking Algorithm, Exposing Harmful Content to Viewers over Months

Oct 1, 2021

Facebook's internal report revealed a six-month long alleged software bug, allowing moderator-flagged posts and other harmful content to byp...

Amazon's AI Cameras Incorrectly Penalized Delivery Drivers for Mistakes They Did Not Make

Sep 20, 2021

A recent incident involving Amazon's AI-driven performance evaluation system incorrectly penalized delivery drivers for errors they did not...

Chinese Province Developed System Tracking Journalists and International Students

Sep 17, 2021

The government of Henan province reportedly developed a system using facial recognition cameras linked to regional and national databases, a...

Apple’s AVs Reportedly Struggled to Navigate Streets in Silicon Valley Test Drives

Sep 1, 2021

Apple's autonomous vehicles experienced difficulties in navigation during on-road tests near their Silicon Valley headquarters. The vehicles...

Startup's Accent Translation AI Denounced as Reinforcing Racial Bias

Aug 15, 2021

A startup's use of AI voice technology to alter or remove accents for call center agents has come under scrutiny as perpetuating racial bias...

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.