AI translation is jeopardizing Afghan asylum claims
Jun 20, 2020
A Pashto-speaking refugee's asylum claim was denied due to a discrepancy between oral and written recount of an event. The error? An automat...
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
TAIM catalog Map Data source & citationUber Allegedly Violated GDPR by Failing to Provide Sufficient Notice on Automated Profiling for Drivers
Jun 20, 2020
Uber is under accusation in a lawsuit for insufficient disclosure about automated decision-making and profiling practices affecting their dr...
Read moreImage Upscaling Algorithm PULSE Allegedly Produced Facial Images with Caucasian Features More Often
Jun 20, 2020
The image upscaling algorithm, PULSE (Powered by NVIDIA's StyleGAN), has been under scrutiny for its perceived preference towards generating...
Read moreiGPT, SimCLR Learned Biased Associations from Internet Training Data
Jun 17, 2020
Uncovering racial, gender, and intersectional biases in unsupervised image generation models like iGPT and SimCLR, which lead to stereotypic...
Read moreStarship Delivery Robot Scuffed Bumper of a Resident’s Car in Texas, Allegedly Refusing to Release Footage of the Accident
Jun 15, 2020
A Starship food delivery robot reportedly collided with a vehicle at a stoplight intersection in Frisco, Texas. The company is accused of re...
Read moreFake Accounts Using GAN Faces Deployed by Propaganda Campaign on Social Platforms
Jun 13, 2020
Pro-China propaganda campaign exploits Generative Adversarial Networks (GANs) to create synthetic faces and deploy fake accounts on platform...
Read moreFacebook’s Moderation Algorithm Banned Users for Historical Evidence of Slavery
Jun 11, 2020
Facebook's automated content moderation system mistakenly censored and banned Australian users for posting an 1890s photo of Aboriginal men...
Read moreImages Captured by iRobot's Roomba Containing Device Users Posted on Private Online Groups
Jun 10, 2020
Private images captured by iRobot's Roomba J7 robot vacuum in user homes, obtained for research purposes with consent, were shared on closed...
Read moreKristen Bell Deepfaked in Non-Consensual AI-Generated Pornography
Jun 8, 2020
Actress Kristen Bell fell victim to a non-consensual deepfake, as her likeness was used in pornographic content without her permission. This...
Read moreMicrosoft’s Algorithm Allegedly Selected Photo of the Wrong Mixed-Race Person Featured in a News Story
Jun 6, 2020
A recent news story on MSN.com mistakenly featured a photo of the wrong mixed-race individual, allegedly chosen by an algorithm. This incide...
Read moreProctorU’s Identity Verification and Exam Monitoring Systems Provided Allegedly Discriminatory Experiences for BIPOC Students
Jun 1, 2020
An identity verification and exam monitoring service used by the University of Toronto, ProctorU, was accused by its students of providing a...
Read moreTesla on Autopilot Crashed into Flipped Truck on Taiwan Highway
Jun 1, 2020
A Tesla Sedan, in Autopilot mode, collided with an overturned truck on a Taiwanese highway, narrowly avoiding a pedestrian. This incident un...
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.