GPT-3-Based Twitter Bot Hijacked Using Prompt Injection Attacks
Sep 15, 2022
A recent incident involving Remoteli.io's GPT-3-based Twitter bot showcased the vulnerability of AI systems to prompt injection attacks. Hac...
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 & citation Failure Files™Delivery Robot Rolled Through Crime Scene
Sep 13, 2022
An autonomous delivery robot, manufactured by Serve Robotics, was filmed traversing a crime scene barricaded by police tape. This incident u...
Read more"The Little Mermaid" Clip Doctored Using Generative AI to Replace Black Actress with White Character
Sep 13, 2022
A disturbing incident involving the doctored clip of Disney's 2022 version of 'The Little Mermaid' has surfaced. The clip, reportedly modifi...
Read moreMisinformation Reported in TikTok's Search Results Despite Moderation by AI and Human
Sep 11, 2022
TikTok's search results have been found to disseminate misleading information on political topics, bypassing both artificial intelligence (A...
Read moreFraudsters Use Deepfake Video of Tim Cook to Promote Apple Crypto Scam on YouTube
Sep 7, 2022
Fraudsters manipulated an old interview with Apple CEO Tim Cook, utilizing AI or video editing, to advertise a false crypto event on YouTube...
Read moreGAN Artwork Won First Place at State Fair Competition
Aug 29, 2022
A digital artwork created using Generative Adversarial Networks (GAN) has emerged victorious in the Colorado State Fair's art competition, s...
Read moreBlenderBot 3 Cited Dutch Politician as a Terrorist
Aug 25, 2022
Meta's conversational AI, BlenderBot 3, mistakenly associated an incumbent Dutch politician with terrorism when prompted about terrorists. T...
Read moreStable Diffusion Exhibited Biases for Prompts Featuring Professions
Aug 22, 2022
AI model Stable Diffusion reportedly exhibited biases in responses to prompts related to professions, raising concerns about gender and cult...
Read moreGoogle’s Automated Child Abuse Detection Wrongfully Flagged a Parent’s Naked Photo of His Child
Aug 21, 2022
Google's automated child safety system, aiming to combat online child abuse, inadvertently flagged a parent's photo intended for a healthcar...
Read moreStable Diffusion Abused by 4chan Users to Deepfake Celebrity Porn
Aug 17, 2022
An open-source image generation model, Stable Diffusion by Stability AI, was reportedly leaked on 4chan before its release date, raising con...
Read moreTesla Autopilot Misidentified On-Road Horse-Drawn Carriage
Aug 15, 2022
Recently, a video surfaced showing Tesla's Autopilot system misinterpreting a horse-drawn carriage as various vehicles, including trucks, ca...
Read moreCFPB Reportedly Finds Hello Digit's Automated Savings Algorithm Caused Overdrafts and Orders Redress with $2.7M Penalty
Aug 10, 2022
The automated savings product of Hello Digit allegedly employed a proprietary algorithm that reportedly led to overdrafts and bank fees, con...
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.