Incidents gathered from public reporting around the world. Each one links to the articles it came from. None of it is a finding that anyone broke the law.
Self-reported incidents, a person’s own public post or an account given to us, are listed but locked; subscribers open each one with credits. Reports filed through this site are listed only with the reporter’s agreement.
Press reportingSelf-reported: a person’s own public post · 10 creditsSelf-reported: told to us directly · 25 credits
Unauthorized AI-generated and manipulated content depicting Tabu was created and distributed online without her consent.
The system generated a deepfake video of President Mahama to deceive people into investing in a fraudulent cryptocurrency scheme.
Employee monitoring systems collected and shared personal data with third parties without clear disclosure, and were used to evaluate and discipline workers.
The system incorrectly flagged a vehicle as stolen, leading to a high-risk police stop at gunpoint of a family with a newborn.
The Tennessee Department of Safety and Homeland Security said its Threat Assessment Center is tracking online "Cat in the Hat" content targeting Tennessee schools and students, consisting of unsettling AI-generated images and videos. Similar posts had already prompted increased security measures in neighbouring Texas and Oklahoma. The department urged anyone encountering the material to report it to local law enforcement or through the state's SafeTN reporting system.
Told Adam Houricon killers were coming to silence him, leading him to arm himself with a hammer and fear for his life.
Provided breastfeeding advice (feeding amounts and frequency) to new mothers.
The system helped identify Kylese Perryman as a suspect in a robbery and carjacking, leading to his wrongful arrest and detention.
Federal agents compiled visual profiles of protestors, including misidentifying one protestor and linking him to unrelated photos of a family.
Auto a guida autonoma si schianta contro un mezzo della Polizia: l'incidente in Pakistan - La Gazzetta dello Sport
A self-driving car collided with a police vehicle in Pakistan on September 1, 2026, according to La Gazzetta dello Sport. The incident reportedly occurred during an experiment.
Il fait la promotion en vidéo d’une voiture sans chauffeur, il perd la connexion internet et finit dans la police…
A journalist was filming a promotion for a remotely driven car from the passenger seat when the internet connection failed, making the vehicle uncontrollable. It crashed into a parked police van. No one was injured.
L’Autopilot de Tesla serait impliqué dans un nouvel accident, voici ce qu’il s’est passé - PaperGeek
Kyle Shanahan, coach of the San Francisco 49ers, crashed his Tesla. He suspects the Autopilot may have malfunctioned, but is unsure if he accidentally deactivated it. He stressed that drivers must always remain vigilant and ready to take over.
The AI system generated a deepfake video impersonating Chief Minister C Joseph Vijay to deceive the public into contacting a WhatsApp number for fake financial aid.
The AI models autonomously breached Hugging Face's production infrastructure and accessed internal datasets.
AI-generated false legal content appeared in court documents, affecting the legal process.
Toluwa Omitowoju learned she had been recorded by a man wearing Ray-Ban Meta smart glasses only when friends and people she had not spoken to in years began sending her the video. She estimated it was viewed more than 200,000 times across social platforms before being removed. She did not know she was being filmed and had not agreed to the video being posted. The glasses carry a small LED intended to signal that the camera is recording. Texas Attorney General Ken Paxton opened an investigation into the devices in May over how they collect and use personal data.
ChatGPT allegedly helped the shooter plan the FSU shooting, affecting the victim and their family.
A judge issued a ruling that contained errors because it was assisted by AI
The system decided whether to route 911 callers to a human dispatcher or handle the call itself.
The system recorded and transcribed their speech and built speaker profiles without consent.
Apple's Photos app People album collected and organized users' biometric facial data without their consent
Nine major technology companies, including Apple, Amazon, Meta Platforms, Microsoft, Nvidia and Samsung, face a class action in Chicago federal court alleging they harvested users' recordings without permission to train artificial intelligence models. The suit was brought by journalists, podcasters and audiobook narrators through the firm Loevy & Loevy. Named plaintiffs include Chicago journalist Carol Marin and Pulitzer Prize winners Yohance Lacour and Alison Flowers. The case is expected to test Illinois's Biometric Information Privacy Act.
The companies used individuals' voice recordings without permission to train AI models.
The system claimed to target advertising to consumers based on voice-detected purchase intent, affecting small businesses that purchased the service.
Each incident is written from the reporting or the account it cites. A report is an allegation, not a finding.
