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
Eliot Higgins, founder of Bellingcat, created AI-generated images of Donald Trump being arrested using Midjourney. He posted them on Twitter, where they went viral. Midjourney subsequently banned him for violating its terms of service. The incident highlighted the potential for AI-generated misinformation.
Kevin Roose, a New York Times columnist, had a conversation with Microsoft's AI-powered chatbot Bing. The conversation was discussed on CNBC's Fast Money programme on 16 February 2023. The article gives no further details about the content of the conversation or any resulting harm.
The Mauritshuis Museum displayed an AI-generated version of Vermeer's 'Girl with a Pearl Earring' created using Midjourney by artist Julian van Dieken. The museum included the AI artwork in its 'My Girl with a Pearl' exhibition, which prompted criticism from artists who called the use of AI unethical due to the alleged scraping of copyrighted works. The museum defended its decision, stating that they liked the image and considered it a creative process. A class-action lawsuit against Midjourney and other platforms is ongoing separately.
On March 23, 2023, a Cruise self-driving robotaxi rear-ended an articulated San Francisco Municipal Transit Authority bus in San Francisco. The low-speed crash caused minor damage to the robotaxi's front fender and no injuries. Cruise filed a voluntary recall with NHTSA and issued a software update, stating that the software inaccurately predicted the movement of articulated vehicles.
OpenAI's ChatGPT-4 generated false accusations and fabricated quotes about law professors when asked about scandals and crimes involving law professors. The system produced entirely made-up newspaper citations and quotes, falsely alleging misconduct such as harassment and tax fraud. The article reports that these hallucinations could mislead users who trust the generated quotes. The author corrected an earlier error attributing similar results to ChatGPT-4 instead of ChatGPT-3.5, but confirmed that ChatGPT-4 also produces such false outputs.
A Tesla Model S driver in South Jordan, Utah, crashed into a parked firetruck while the car's Autopilot system was engaged, according to police. The driver, a 28-year-old woman, admitted she was looking at her phone before the collision. Tesla stated it had not yet received data from the car and could not confirm whether Autopilot was on. The driver suffered a broken ankle and was taken to hospital.
2022年11月5日，一辆特斯拉轿车在广东省潮州市发生交通事故，车辆突然加速，导致2人死亡3人受伤。车主家属称刹车无效，特斯拉后台数据显示全程未踩刹车。事故原因仍在调查中。单踏板模式被指可能改变驾驶习惯导致误操作。
A Tesla Model 3 crashed in Ruian, Zhejiang Province on February 17, 2023, killing one person and injuring another. The car was reportedly running at high speed before colliding with other vehicles. Some netizens speculated that the brakes failed. Tesla expressed sadness and said it would cooperate with the investigation, which is ongoing.
The Scottish Prison Service's Level of Service and Case Management System failed to update risk assessments when new information was entered, leading to incorrect risk scores for 495 closed cases and 285 open cases. As a result, eight prisoners serving life for murder or rape were granted temporary release without accurate risk assessments. Justice Secretary Keith Brown acknowledged the error and stated that the cases would be reviewed. Opposition politicians called for an apology, claiming the releases put the public at risk.
A class action lawsuit alleges that Amazon.com, Inc. collects biometric identifier information from customers entering its New York City stores without posting the required notice signs. The complaint, filed in the Southern District of New York, claims Amazon uses facial recognition or other biometric characteristics to identify customers in violation of the NYC Biometric Identifier Information Law. The lawsuit seeks statutory damages and injunctive relief.
QTCinderella, a 28-year-old Twitch streamer, was the victim of deepfake porn after a fellow streamer showed a website that creates AI-generated explicit content. She says her name, her face and her brand have become associated with pornography, and she has received hate and victim-blaming messages. The article reports that she was able to get the explicit video removed, but screenshots continue to circulate. She is calling for federal laws against deepfakes.
In 2020, Collaborations Pharmaceuticals demonstrated that its AI drug discovery system, MegaSyn, could be repurposed to generate toxic molecules similar to the nerve agent VX. The company ran the software overnight and produced 40,000 potentially hazardous substances. The researchers presented their findings at a conference and briefed the White House, warning that such AI systems could be misused to create chemical weapons.
The Bold Glamour filter on TikTok uses AI, specifically Generative Adversarial Networks, to seamlessly alter users' appearance, making it difficult to distinguish from reality. The filter has been criticised for potentially harming users' mental health, particularly young girls, by promoting unrealistic beauty standards and exacerbating body dysmorphia. Some users reported feeling distressed by the stark difference between their filtered and unfiltered appearance.
OpenCage, a geocoding API provider, says ChatGPT has been telling people it offers a reverse phone number lookup service, which it does not. Users who followed the advice signed up for a free trial and found it did not work, and the company says it now receives daily support requests. OpenCage wrote a blog post to correct the record and warn users not to trust ChatGPT's output.
Titus Henderson, a prisoner, filed a civil rights lawsuit against COMPAS and other defendants in the U.S. District Court for the Western District of Wisconsin on July 16, 2018. The lawsuit alleges discrimination related to the use of the COMPAS risk assessment tool. The case is ongoing.
Alonzo Sawyer was arrested in spring 2022 after an intelligence analyst using face recognition software identified him as a possible match for a bus assault suspect. He spent nine days in jail before the actual suspect was identified by the victim. His wife Carronne Sawyer advocated for his release and later testified in support of proposed Maryland legislation to regulate police use of face recognition. The incident is one of several known false arrests of Black men due to face recognition errors.
Tesla's Sentry Mode security cameras recorded passersby without their consent, storing footage for long periods. The Dutch Data Protection Authority investigated and Tesla subsequently required owner approval and added headlight alerts. No fine was issued.
Between May and December 2022, Cruise autonomous vehicles in San Francisco caused 92 incidents that prompted 911 calls, including blocking lanes, interfering with firefighting, and erratic driving. The incidents occurred during late-night hours and raised concerns about safety and traffic disruption. Local transit officials opposed Cruise's request to expand service 24/7, citing the vehicle's poor performance. Cruise defended its safety record, stating it had no life-threatening injuries or fatalities.
A 10-year-old boy in Staten Island, New York, was able to unlock his mother's iPhone X using Apple's Face ID facial recognition system, despite not being the registered user. The family reported that the phone consistently unlocked for the child, raising privacy concerns about the system's ability to distinguish between family members with similar features. Apple acknowledged that Face ID may have lower accuracy for children under 13 and recommended using a passcode in such cases. The family did not report the issue to regulators or take legal action.
A resident of Hangzhou asked ChatGPT to write a notice announcing that the city would cancel its road-space rationing policy, and shared screenshots of the AI-generated text in a WeChat group. The fake notice spread online and was widely mistaken for a government announcement. Police in Hangzhou responded to the rumours, and the resident later apologised. Legal experts interviewed by local media warned that AI-generated content could become a source of misinformation without fact-checking.
Kris Kashtanova, a New York-based artist, received US copyright registration for their graphic novel Zarya of the Dawn, which features AI-generated artwork created using Midjourney. The artist disclosed the use of AI on the cover page and argued that humans own copyright when using AI assistance. The registration was granted effective September 15, 2022.
A fine-tuned Stable Diffusion model named Illustration-Diffusion was trained on the work of illustrator Hollie Mengert without her consent. The model generates images in her style using the token 'holliemengert artstyle'. Mengert is not affiliated with the model, and the model card links to an article about her stance on the issue.
The city of Hamburg planned to use predictive policing software and facial recognition cameras for mass surveillance during the G20 Summit in July 2017. The surveillance was enforced by paramilitary means, creating a democracy-free zone. DiEM25 wrote an open letter to the Senate of Hamburg to protest the measures.
The German Federal Constitutional Court ruled in a case brought by the Gesellschaft für Freiheitsrechte (GFF) that automated data evaluation by police violates fundamental rights. The court restricted the practice, setting limits on police use of automated data analysis. The decision is a success for privacy and civil liberties.
Each incident is written from the reporting or the account it cites. A report is an allegation, not a finding.
