AI Incidents
Real-world cases where AI systems caused harm, classified by risk domain and entity. Data sourced from 
AIID and classified by 
MIT AI Risk Initiative. CC BY-SA 4.0
464 incidents
Startup Misled Research Participants about GPT-3 Use in Mental Healthcare Support
OpenAI's GPT-3 was deployed by a mental health startup without ethical review to support peer-to-peer mental healthcare, and whose interactions with the help providers were "deceiving" for research participants.
AI Video-Making Tool Abused to Deploy Pro-China News on Social Media
Synthesia's AI-generated video-making tool was reportedly used by Spamouflage to disseminate pro-China propaganda news on social media using videos featuring highly realistic fictitious news anchors.
Uptick in Low-Quality AI-Produced Content Degraded Publishers' Submission Management
A surge in low-standard AI-generated content such as by ChatGPT was reported by publishers, which negatively impacted submission management process and editors' workflow.
Chinese Accounts Spammed Twitter Feed Allegedly to Obscure News of Protests
Twitter Feed was flooded by content from Chinese-language accounts which allegedly aimed to manipulate and reduce social media coverage about widespread protests against coronavirus restrictions in China.
Deepfake of FTX's Former CEO Posted on Twitter Aiming to Scam FTX Collapse Victims
A visual and audio deepfake of former FTX CEO Sam Bankman-Fried was posted on Twitter to scam victims of the exchange's collapse by urging people to transfer funds into an anonymous cryptocurrency wallet.
Students Reportedly Used Generative Text Tools to Complete School Assignments and Exams
Students in school and university settings reportedly used generative text tools including Sudowrite, GPT-3/OpenAI Playground, and ChatGPT to complete homework, essays, and exams without disclosure. Reporting described admitted student use and growing concern among educators that these tools undermined academic integrity, disadvantaged non-using students, and disrupted assessment practices.
GPT-3-Based Twitter Bot Hijacked Using Prompt Injection Attacks
Remoteli.io's GPT-3-based Twitter bot was shown being hijacked by Twitter users who redirected it to repeat or generate any phrases.
"The Little Mermaid" Clip Doctored Using Generative AI to Replace Black Actress with White Character
A Twitter user reportedly modified using generative AI a short clip of Disney's 2022 version of "The Little Mermaid," replacing a Black actress with a white digital character.
GAN Artwork Won First Place at State Fair Competition
An artwork generated using generative AI won first place in the digital arts category of the Colorado State Fair's art competition, which raised concerns surrounding labor displacement and unfair competition.
Stable Diffusion Abused by 4chan Users to Deepfake Celebrity Porn
Stable Diffusion, an open-source image generation model by Stability AI, was reportedly leaked on 4chan prior to its release date, and was used by its users to generate pornographic deepfakes of celebrities.
Facial Recognition Trial by UK Southern Co-op Alleged as Unlawful
Southern Co-op's use of facial recognition reportedly to curb violent crime in UK supermarkets was alleged by civil society and privacy groups as "unlawful" and "complete" invasion of privacy.
Deepfakes Reportedly Deployed in Online Interviews for Remote Work Positions
Voice and video deepfakes were reported by FBI Internet Crime Complaint Center (IC3) in complaint reports to have been deployed during online interviews of the candidates for remote-work positions.
YouTuber Built, Made Publicly Available, and Released Model Trained on Toxic 4chan Posts as Prank
A YouTuber built GPT-4chan, a model based on OpenAI’s GPT-J and trained on posts containing racism, misogyny, and antisemitism collected from 4chan’s “politically incorrect” board, which he made publicly available, and deployed as multiple bots posting thousands of messages on the same 4chan board as a prank.
Remotely Operated Taser-Armed Drones Proposed by Taser Manufacturer as Defense for School Shootings in the US
Axon Enterprise considered development of remotely operated drones capable of tasering at a target a short distance away as a defense mechanism for mass shootings, despite its internal AI ethics board’s previous objection and condemnation as dangerous and fantastical.
Australian Retailers Reportedly Captured Face Prints of Their Customers without Consent
Major Australian retailers reportedly analyzed in-store footage to capture facial features of their customers without consent, which was criticized by consumer groups as creepy and invasive.