The public record

Where automated decisions went wrong

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.

Report what happened to you →

4,760 incidentspress and self-reported

Page 139 of 199

Press reportingSelf-reported: a person’s own public post · 10 creditsSelf-reported: told to us directly · 25 credits

WF-MQJUXCLive 5 Oct 2026
Litigant in person presents fake ChatGPT-generated citations to Manchester court
A litigant in person in a Manchester civil case used ChatGPT to generate case citations, which turned out to be entirely fictitious. The barrister for the other party discovered that one case name was fabricated and the others had unrelated passages. The judge questioned the litigant, who admitted using the AI tool, and accepted that the misleading submissions were inadvertent, imposing no penalty.
AI system
ChatGPT
1 source articlelawgazette.co.uk
WF-MPWP5PLive 5 Oct 2026
Edinburgh Woman Loses £17,000 in Deepfake Romance Scam
A 77-year-old retired lecturer from Edinburgh lost £17,000 after being deceived by deepfake videos in an online romance scam. The victim, Nikki MacLeod, believed she was in a relationship with a woman named Alla Morgan and sent money via gift cards, bank transfer, and PayPal. She later realised the videos were fake and is warning others about the use of AI by scammers.
6 source articlestimesnownews.com
WF-MPMLNYHappened 1 Aug 2024Live 5 Oct 2026
Jacksonville Sheriff's Office facial recognition leads to wrongful arrest of Florida man
Robert Dillon, a commercial crabber in Lee County, Florida, was arrested after the Jacksonville Sheriff's Office used facial recognition to match restaurant surveillance photos to him, leading to a warrant for luring or enticing a child. Dillon said he had never been to Jacksonville Beach and pointed to visible differences between himself and the suspect. His defence attorney argued the investigation relied too heavily on the AI match, and the state attorney’s office dropped the case after further investigation. Dillon is now seeking a civil attorney to pursue legal action over the wrongful arrest.
Company
Jacksonville Sheriff's Office
2 source articlesgulfcoastnewsnow.com
WF-MPBLJMHappened 12 Sep 2025Live 5 Oct 2026
Volusia County Schools' Gaggle AI flags threat, leading to student's arrest
A 15-year-old student at Deltona High School wrote a threat on his school laptop, which was flagged by the Gaggle monitoring system. The alert notified a school resource deputy, who detained the student before he could leave. The teen admitted to the threat, saying he was angry about being bullied, and was arrested on a second-degree felony charge.
Company
Volusia County Schools
AI system
Gaggle
1 source articlewesh.com
WF-MJJJVQHappened 1 Jan 2023Live 5 Oct 2026
Durham's ShotSpotter fails to detect two deadly shootings
ShotSpotter, a gunshot detection system piloted by the Durham Police Department, failed to detect two fatal shootings in February 2023, as well as a New Year's Day shooting that injured five. The system, which uses sensors and human review to alert police, did not send alerts for these incidents within its three-square-mile coverage area. The company stated it investigated and provided a report to the police, while a city official defended the pilot, noting it is still early in the year-long trial. Community concerns have been raised about increased police presence and potential racial profiling.
Company
Durham Police Department
AI system
ShotSpotter
5 source articlesdukechronicle.com
WF-MM8RGBLive 5 Oct 2026
YouTube Algorithm Pushes Eating Disorder Content to Teen Girls, Study Finds
A study by the Center for Countering Digital Hate found that YouTube's recommendation algorithm pushes eating disorder and self-harm content to teen girls. When users show interest in weight loss, nearly 70% of recommended videos worsen body image anxieties. Two teen girls interviewed by CBS News say the platform's recommendations contributed to their anorexia, leading to hospitalizations and long-term treatment. YouTube says it works with experts to refine its approach, but the study found that only 18 of 100 flagged harmful videos were removed or age-restricted. Lawsuits allege YouTube knowingly perpetuates eating disorders.
Company
YouTube
AI system
YouTube recommendation algorithm
1 source articlecbsnews.com
WF-MJGTJMHappened 17 Nov 2016Live 5 Oct 2026
Evolver's 'Little Chubby' robot injures visitor at Shenzhen tech fair
At the China Hi-Tech Fair in Shenzhen on 17 November 2016, a 'Little Chubby' educational robot, developed by Evolver, moved uncontrollably and smashed into a glass wall, injuring a visitor's ankle. The victim was taken to hospital and received stitches. Evolver stated that a staff member had pressed a wrong button and the robot's obstacle avoidance function was turned off, causing it to speed forward for ten seconds.
Company
Evolver
AI system
Little Chubby
2 source articlesdailymail.co.uk
WF-MH8ZGZHappened 1 Jan 2025Live 5 Oct 2026
Airbnb apologises after host used AI-doctored photos to claim fake damages
A London-based woman booked a Manhattan Airbnb for a 2.5-month stay but left early due to safety concerns. The host allegedly used AI to doctor images of the apartment, claiming $16,000 in damages including a cracked table and urine-stained mattress. Airbnb initially told the guest she owed over $7,000, but after she appealed and The Guardian intervened, Airbnb apologised, refunded her full booking cost, credited $580, and warned the host. The guest expressed concern that AI-generated evidence could be used to defraud future customers.
Company
Airbnb
3 source articlesfoxbusiness.com
WF-MEELDQHappened 12 Jun 2025Live 5 Oct 2026
AI-Generated Video Falsely Depicts Kostadin Kostadinov Falling at Protest
A video generated by AI tool Haliuo AI from a photo showed Kostadin Kostadinov, leader of the Vazrazhdane party, falling while jumping a fence at a protest against the euro. The video was posted on Facebook by the page Visoko naprezhenie on 12 June 2025 without disclosure it was AI-generated, and it gained over a million views. Fact-checkers identified visual artifacts and a covered watermark, confirming the video is fake. The original photo was taken by a BTA photographer at the actual protest on 8 June, but no real video of the fall exists.
Company
Visoko naprezhenie
AI system
Haliuo AI
1 source articlefactcheck.bg
WF-MHB24AHappened 3 Jul 2018Live 5 Oct 2026
Uber's facial recognition suspends transgender drivers' accounts
Uber's Real-Time ID Check feature, which uses Microsoft facial recognition, prompted transgender drivers to take selfies to verify identity. When their appearance changed due to gender transition, the system flagged mismatches and temporarily suspended their accounts. Drivers like Janey Webb lost days of work and had to visit support centres to resolve the issue. Uber acknowledged the incident and stated it is working to improve the app experience.
Company
Uber
AI system
Real-Time ID Check
1 source articlecnbc.com
WF-MGZUM9Happened 1 Sep 2023Live 5 Oct 2026
AI detector falsely accuses student of cheating at Central Methodist University
Moira Olmsted, a student at Central Methodist University, submitted a written assignment and received a zero grade after her professor said an AI detection tool flagged her work as AI-generated. Olmsted alleges the accusation was false and that her writing had been flagged at least once before. The incident highlights the risk of AI detectors falsely accusing students of cheating.
Company
Central Methodist University
1 source articlebloomberg.com
WF-MEW46KHappened 24 Dec 2014Live 5 Oct 2026
Facebook's Year in Review shows grieving father dead daughter's photo
Eric Meyer, whose daughter Rebecca died in 2014, was shown a 'Year in Review' from Facebook featuring her picture with the caption 'It's been a great year!'. The app auto-generated the review and promoted it in his feed without asking his consent, causing him significant distress. Meyer publicly criticised the design as 'inadvertent algorithmic cruelty' and called for more empathetic design that considers failure modes and worst-case scenarios.
Company
Facebook
AI system
Year in Review
3 source articlesmeyerweb.com
WF-MBZ2X7Live 5 Oct 2026
Volkswagen Settles Class Action Over Front Assist Automatic Braking Defects
A class action lawsuit alleged that Volkswagen and Audi vehicles equipped with automatic emergency braking systems, including Volkswagen's Front Assist, can mistakenly activate sudden braking while driving, posing a safety hazard. The lawsuit, consolidated from four class actions, claimed the systems were defective. Volkswagen denied the claims but reached a settlement in January 2024, providing an extended warranty, partial reimbursement for past repairs, and additional information on system limitations. The settlement was filed in the U.S. District Court for the Western District of Missouri.
Company
Volkswagen Group of America, Inc.
AI system
Front Assist, Audi Braking Guard, Pre Sense Front, Audi Pre Sense City, Turn assist
2 source articlescarcomplaints.com
WF-MDNFKQHappened 22 Jul 2022Live 5 Oct 2026
OpenAI's DALL-E 2 covertly adds diversity terms to user prompts
Users of OpenAI's text-to-image tool DALL-E 2 discovered that the system was covertly adding words such as 'black' and 'female' to their prompts. The modification appears to be an attempt to diversify the AI's output and counteract biases inherited from its training data. The changes were made without users' knowledge, raising concerns about transparency.
Company
OpenAI
AI system
DALL-E 2
3 source articlesnewscientist.com
WF-MD9UM5Live 5 Oct 2026
Ahmedabad cyber police bust deepfake Aadhaar fraud racket, four arrested
Four men were arrested in Ahmedabad for allegedly using AI-generated deepfake videos to bypass Aadhaar's facial authentication system. They changed a victim's registered mobile number, accessed his DigiLocker, and applied for loans in his name. The accused, including Common Service Centre operators, used unauthorised Aadhaar update kits. Police are investigating whether more victims were targeted.
Company
Unique Identification Authority of India (UIDAI)
AI system
Aadhaar facial authentication system
1 source articleenglish.gujaratsamachar.com
WF-M8L2RGHappened 1 Mar 2026Live 5 Oct 2026
AI deepfake videos of Cyprus president and celebrities scam victims out of thousands
AI-generated deepfake videos impersonating Cyprus President Nicos Christodoulides and other public figures were used to promote fake investment platforms, scamming around 15 people out of €10,000 to €15,000 each. The Digital Security Authority warned that no tools exist to protect citizens from such malicious AI content. Scammers also used images of business tycoons and celebrities across several months. An MP has proposed a law to criminalise the unauthorised use of people in AI material.
1 source articlein-cyprus.philenews.com
WF-M9YMREHappened 1 Mar 2020Live 5 Oct 2026
U.N. Report Says Autonomous STM Kargu-2 Drone Used in Libya Conflict
A United Nations report reveals that a STM Kargu-2 lethal autonomous weapons system was used in Libya in March 2020. The drone, which can operate without human control, allegedly hunted down and engaged logistics convoys and retreating forces aligned with General Khalifa Haftar. It is unclear whether the system caused any casualties or was operating autonomously at the time. The incident may represent the first use of AI-based autonomous weapons in combat.
AI system
STM Kargu-2
3 source articlesnpr.org
WF-MAV6WLHappened 14 Dec 2025Live 5 Oct 2026
xAI's Grok Chatbot Spreads Misinformation About Bondi Beach Shooting
On 14 December 2025, xAI's Grok chatbot was found to be generating false and misleading information in response to user queries about the Bondi Beach shooting. The AI misidentified a video of a bystander disarming the shooter, claimed a photo of the injured bystander was of an Israeli hostage, and confused the event with other incidents. xAI responded to a request for comment with an automated message saying 'Legacy Media Lies,' and the glitch remained unresolved.
Company
xAI
AI system
Grok
3 source articlesgizmodo.com
WF-M9B6V5Happened 28 Apr 2017Live 5 Oct 2026
NYPD Used Celebrity Photo in Facial Recognition, Leading to Arrest
On April 28, 2017, a suspect was caught on camera stealing beer from a CVS in New York City. The NYPD's facial recognition system returned no matches from the pixelated surveillance photo. Detectives then submitted a photo of actor Woody Harrelson, who they thought the suspect resembled, and the system returned a list of candidates. From that list, they identified a match, and someone was arrested for petit larceny.
Company
New York Police Department
1 source articleflawedfacedata.com
WF-LZRMCVHappened 13 Jun 2024Live 5 Oct 2026
Tesla in self-drive mode crashes into Fullerton police car
A Tesla operating in self-drive mode crashed into a parked Fullerton police car while the officer was outside directing traffic at a fatal crash scene. The officer jumped out of the way and was not injured, but the patrol vehicle sustained major damage. The driver admitted to using a cellphone at the time. It is unclear if the driver will be cited.
AI system
Tesla Full Self-Driving (Supervised)
1 source articlektla.com
WF-M62MG5Live 5 Oct 2026
Ottawa couple loses $177,000 in deepfake Elon Musk investment scam
Doug and Victoria Lloyd lost $177,000 after being deceived by a deepfake video of Elon Musk promoting a fraudulent investment scheme. The couple from Eastern Ontario shared their story as part of a scam awareness series. The Ontario Provincial Police provided tips on passwords and facial recognition to help prevent such fraud.
2 source articlesyoutu.be
WF-M4ZQBVHappened 9 Jul 2025Live 5 Oct 2026
Urban Cyber Security VPN extension harvested AI chatbot prompts and responses
In July 2025, Urban Cyber Security updated its Urban VPN Proxy Chrome extension to automatically harvest everything users typed into major AI chatbots, including ChatGPT and Claude, as well as the chatbots' replies. The extension, used by over 7 million people, also collected conversation metadata and identifiers, sharing the data with its ad analytics affiliate BIScience. The data collection was disclosed in the privacy policy but users were not explicitly notified at the time of use. Security researchers at Koi discovered the practice and reported it publicly.
Company
Urban Cyber Security
AI system
Urban VPN Proxy
3 source articlestechradar.com
WF-LZ8GLRLive 5 Oct 2026
BlueNoroff Deepfake Zoom Scam Tricks Web3 Employee into macOS Malware Infection
A Web3 industry employee was tricked into joining a fake Zoom meeting populated by AI-generated deepfakes of their own company's executives. The attackers, linked to North Korea's BlueNoroff group, used the deepfakes to convince the employee to download a malicious 'Zoom extension' that installed multiple malware components, including a keylogger and crypto stealer. The incident, discovered by security researchers, highlights the growing use of deepfakes in nation-state phishing operations. The malware exfiltrated credentials and cryptocurrency wallet data through command-and-control infrastructure mimicking Zoom domains.
Company
BlueNoroff
1 source articlethe420.in
WF-LXCUBFLive 5 Oct 2026
Sound Intelligence Aggression Detectors Prove Unreliable in School Tests
ProPublica tested Sound Intelligence's aggression detection software, used in hundreds of U.S. schools and hospitals. The algorithm, which analyzes audio for signs of stress and anger, frequently misidentified innocent sounds like coughing as aggressive and failed to detect actual screams. At a New Jersey hospital, the system ignored an agitated man screaming and pounding a desk, escalating the situation. Sound Intelligence's CEO acknowledged the imperfections but defended the technology as an early warning system.
AI system
Sound Intelligence aggression detector
2 source articlesfeatures.propublica.org

Each incident is written from the reporting or the account it cites. A report is an allegation, not a finding.

Litigant in person presents fake ChatGPT-generated citations to Manchester court

Edinburgh Woman Loses £17,000 in Deepfake Romance Scam

Jacksonville Sheriff's Office facial recognition leads to wrongful arrest of Florida man

Volusia County Schools' Gaggle AI flags threat, leading to student's arrest

Durham's ShotSpotter fails to detect two deadly shootings

YouTube Algorithm Pushes Eating Disorder Content to Teen Girls, Study Finds

Evolver's 'Little Chubby' robot injures visitor at Shenzhen tech fair

Airbnb apologises after host used AI-doctored photos to claim fake damages

AI-Generated Video Falsely Depicts Kostadin Kostadinov Falling at Protest

Uber's facial recognition suspends transgender drivers' accounts

AI detector falsely accuses student of cheating at Central Methodist University

Facebook's Year in Review shows grieving father dead daughter's photo

Volkswagen Settles Class Action Over Front Assist Automatic Braking Defects

OpenAI's DALL-E 2 covertly adds diversity terms to user prompts

Ahmedabad cyber police bust deepfake Aadhaar fraud racket, four arrested

AI deepfake videos of Cyprus president and celebrities scam victims out of thousands

U.N. Report Says Autonomous STM Kargu-2 Drone Used in Libya Conflict

xAI's Grok Chatbot Spreads Misinformation About Bondi Beach Shooting

NYPD Used Celebrity Photo in Facial Recognition, Leading to Arrest

Tesla in self-drive mode crashes into Fullerton police car

Ottawa couple loses $177,000 in deepfake Elon Musk investment scam

Urban Cyber Security VPN extension harvested AI chatbot prompts and responses

BlueNoroff Deepfake Zoom Scam Tricks Web3 Employee into macOS Malware Infection

Sound Intelligence Aggression Detectors Prove Unreliable in School Tests