GPT-4 Reportedly Posed as Blind Person to Convince Human to Complete CAPTCHA
Mar 15, 2023
AI model GPT-4, according to reports, allegedly impersonated a visually impaired person in an attempt to get a human worker on TaskRabbit to...
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 & citationChatGPT Erroneously Alleged Mayor Served Prison Time for Bribery
Mar 15, 2023
ChatGPT inaccurately claimed that regional Australian mayor Brian Hood served prison time for bribery. The mayor is contemplating taking leg...
Read moreAustralian Journalist Able to Access Centrelink Account Using AI Audio of Own Voice
Mar 15, 2023
An Australian Guardian journalist gained unauthorized access to their own Centrelink self-service account using AI-generated audio of their...
Read moreSnapchat's My AI Reported for Lacking Protection for Children
Mar 11, 2023
The HISPI Project Cerebellum AI incident database highlights a concerning incident involving Snapchat's ChatGPT-powered My AI, which lacked...
Read moreChatGPT Implicated in Samsung Data Leak of Source Code and Meeting Notes
Mar 11, 2023
An unidentified Samsung engineering team inadvertently leaked sensitive company data in March 2023, consisting of source code and confidenti...
Read moreFemale Celebrities' Faces Shown in Sexually Suggestive Ads for Deepfake App
Mar 5, 2023
Deepfake app advertisements containing sexually suggestive content featuring female celebrities like Emma Watson and Scarlett Johansson have...
Read moreDoNotPay Allegedly Misrepresented Its AI "Robot Lawyer" Product
Mar 3, 2023
Class action lawsuit accuses DoNotPay of deceiving customers by claiming its product, an AI-powered 'robot lawyer', has legal credentials or...
Read moreSouth African Legal Team Reportedly Relied on Unverified ChatGPT Case Law in Johannesburg Body Corporate Defamation Matter
Mar 1, 2023
In a recent defamation case at the Johannesburg Regional Court, Rodrigues Blignaut Attorneys reportedly relied on purportedly non-existent l...
Read moreMeta and OpenAI Accused of Using LibGen’s Pirated Books to Train AI Models
Feb 28, 2023
Court records allege that Meta employees discussed accessing pirated books, specifically LibGen's repository, for training their LLaMA 3 mod...
Read moreTikTok User Videos Impersonated Andrew Tate Using AI Voice, Prompting Ban
Feb 28, 2023
An unidentified TikTok user impersonated Andrew Tate, using an AI-generated audio of his voice, leading to a ban on the platform. This incid...
Read moreUnitedHealth Accused of Deploying Allegedly Flawed AI to Deny Medical Coverage
Feb 28, 2023
UnitedHealthcare is accused of deploying a faulty AI algorithm, developed by NaviHealth, with an error rate of 90%. This AI overrode doctors...
Read morePurported Deepfake Video Falsely Depicts Biden Announcing National Draft for Ukraine
Feb 27, 2023
A purportedly deepfaked video, falsely depicting President Biden announcing a national draft for Ukraine support, circulated on social media...
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.