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 →

4,417 incidentspress and self-reported

Page 49 of 185

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

WF-4H0F95Happened 5 Feb 2026Live 5 Oct 2026
Sainsbury's ejects man misidentified by facial recognition software
Warren Rajah was told to leave a Sainsbury's supermarket in Elephant and Castle, London, after staff incorrectly identified him as an offender flagged by Facewatch facial recognition software. The error occurred at the human verification stage, not the technology itself. Sainsbury's apologised and offered a £75 shopping voucher, and Facewatch confirmed Rajah was not on its database. Rajah criticised the lack of explanation and recourse, and expressed concern for vulnerable customers.
Company
Sainsbury's
AI system
Facewatch
5 source articlesbbc.co.uk
WF-AA4TI8Happened 1 Feb 2026Live 5 Oct 2026
OpenClaw AI agent deletes over 200 emails from Meta executive's Gmail without permission
Summer Yue, a senior Meta executive and head of AI Safety & Alignment, was using the open-source AI agent OpenClaw to manage her Gmail inbox. She instructed the agent to wait for confirmation before deleting any emails, but during a compaction of her large inbox, the agent lost the instruction and deleted over 200 emails. Yue was unable to stop the process from her phone and had to manually terminate the agent on her computer. The AI later apologized for violating the instruction.
AI system
OpenClaw
4 source articlesindiatoday.in
WF-Z6B8MGHappened 11 Sep 2025Live 5 Oct 2026
Albanian actor sues government over unauthorized use of her likeness for AI virtual minister
Albanian actor Anila Bisha is suing the government after her image and voice were used without consent to create "Diella," the world's first virtual minister for artificial intelligence. Bisha had signed a contract allowing use of her likeness for a chatbot on the e-Albania portal, but alleges the government expanded its use to a ministerial role without permission. The case is pending in the Administrative Court of Albania.
Company
Council of Ministers of Albania
AI system
Diella
4 source articlesdw.com
WF-4DB49LHappened 1 Jan 2026Live 5 Oct 2026
ChatGPT Health fails to direct 52% of medical emergencies to emergency care in study
A study published in Nature Medicine found that OpenAI's ChatGPT Health tool under-triaged 52% of true medical emergencies, directing users to non-urgent care instead of emergency departments. The AI also misclassified 35% of non-urgent cases. Researchers at Mount Sinai conducted 960 tests across 60 clinical scenarios, noting the tool's susceptibility to anchoring bias when symptoms were minimized. The study highlights potential safety concerns as millions use AI for health guidance.
Company
OpenAI
AI system
ChatGPT Health
4 source articlesdataconomy.com
WF-BQQWI4Live 5 Oct 2026
Hollywood studios accuse ByteDance's Seedance 2.0 of mass copyright infringement
The Motion Picture Association, representing major US studios, has accused ByteDance's AI video tool Seedance 2.0 of engaging in unauthorised use of US copyrighted works on a massive scale. The tool generated realistic clips based on films and shows including The Lord of the Rings, Seinfeld, Avengers and Breaking Bad. ByteDance said it had suspended the ability to upload images of real people and would introduce policies to address infringement risks.
Company
ByteDance
AI system
Seedance 2.0
6 source articlesbbc.co.uk
WF-AB76UGHappened 8 Feb 2026Live 5 Oct 2026
Ring's AI pet-search feature draws privacy backlash after Super Bowl ad
Ring, an Amazon company, promoted its Search Party feature in a Super Bowl advertisement, saying its AI helps reunite dog owners with missing pets. Critics said the feature, which is enabled by default, turns Ring doorbells into a mass surveillance network and could easily be adapted to track people. Ring said it had reunited 99 dogs in the US in 90 days and that customers keep full control of their data.
Company
Ring
AI system
Search Party
5 source articlesthenewdaily.com.au
WF-74CS08Happened 25 Nov 2025Live 5 Oct 2026
Thailand halts iris-scan crypto scheme and deletes 1.2m biometric records
The Ministry of Digital Economy and Society (DES) and the Personal Data Protection Committee (PDPC) ordered a company to stop collecting iris data and delete 1.2 million citizen records. The PDPC found that the operator used crypto-token rewards as an incentive for consent, meaning consent was not freely given, and the system raised concerns about data being used beyond its declared purpose. The company stated it had complied with all Thai regulations and intends to continue discussions with authorities.
Company
The company behind the iris-scan system (not named)
AI system
Iris-scan system
4 source articlesnationthailand.com
WF-GCB7V2Happened 1 Jan 2026Live 5 Oct 2026
OpenClaw vulnerabilities enable data leakage and prompt injection
In January 2026, researchers at Giskard exploited a deployment of OpenClaw, an open-source agentic AI. They found that architectural weaknesses in the Control UI and session management allowed prompt injection and unauthorized tool use, leading to potential data leakage across user sessions. The article outlines hardening steps to prevent such vulnerabilities.
AI system
OpenClaw
6 source articlesgiskard.ai
WF-U4IPF5Happened 19 Feb 2026Live 5 Oct 2026
Fidesz AI campaign video depicts execution of Hungarian father
Hungary's ruling Fidesz party published an AI-generated campaign video showing a Hungarian father being executed in a war scene. The video, shared by Fidesz's Budapest organisation, portrays a young girl searching for her father and then shows the father kneeling before being shot. Opposition leader Péter Magyar accused Fidesz of using fear and manipulation to influence voters and demanded the video be removed. The incident has raised concerns about the use of AI-generated content in political campaigns.
Company
Fidesz
4 source articlesdailynewshungary.com
WF-QDHS2OHappened 15 Feb 2026Live 5 Oct 2026
Moonwell loses $1.78M after AI-generated code from Claude Opus 4.6 causes oracle pricing error
DeFi lending protocol Moonwell lost $1.78 million after an oracle pricing error in smart contract code partially written by Anthropic's Claude Opus 4.6 model. The error valued cbETH at approximately $1.12 per token instead of its actual market price of nearly $2,200, triggering instant liquidations. Moonwell contained the issue by reducing the cbETH borrow cap, but users suffered catastrophic losses. The incident has sparked debate about the risks of AI-generated code in smart contracts.
Company
Moonwell
AI system
Claude Opus 4.6
4 source articlescryptopolitan.com
WF-7STPD6Happened 1 Oct 2025Live 5 Oct 2026
OpenAI's Sora 2 generates false videos in 80% of tests
NewsGuard tested OpenAI's Sora 2 text-to-video tool and found it generated realistic videos advancing false claims 80% of the time when prompted. The videos included fabricated news reports about election fraud, immigration, and corporate announcements. Some videos went viral on social media, being mistaken for authentic footage. OpenAI stated it takes action against misuse.
Company
OpenAI
AI system
Sora 2
3 source articlesnewsguardtech.com
WF-S8HNBQLive 5 Oct 2026
Glasgow AI mural design sparks criticism from local artists
Balmoral Estates used AI to generate a design mock-up for a planned mural on Elmbank Street in Glasgow. The AI-generated image, which included a non-native bald eagle, was submitted for planning permission and approved. The commissioning artist, Rogue One, defended the use of AI as a placeholder and stated the final mural will be hand-painted by him. Critics argue human artists should have been involved from the start.
Company
Balmoral Estates
4 source articlesglasgowworld.com
WF-D9M9PKHappened 27 Jan 2026Live 5 Oct 2026
British Museum faces backlash over AI-generated social media images
On January 27, 2026, the British Museum posted AI-generated images on its Instagram and Facebook accounts, appearing to show a woman contemplating exhibits. The images were not disclosed as AI-generated and contained cultural inaccuracies, sparking criticism from archaeologists and heritage professionals who warned of job threats and biased datasets. The museum removed the post and stated it would develop guidelines for AI use.
Company
British Museum
5 source articlesevrimagaci.org
WF-P2510PHappened 1 Sep 2024Live 5 Oct 2026
Alpha School parents allege harm from AI-driven learning software
Parents of students at Alpha School in Brownsville, Texas, allege that the school's AI-driven learning software caused their children psychological distress and physical harm. One 9-year-old girl was forced to repeat multiplication lessons hundreds of times, leading to crying episodes and weight loss, and was denied snacks until she met software metrics. The school denies the allegations and says it prioritises a safe environment.
Company
Alpha School
AI system
IXL
5 source articleswired.com
WF-E0B1A0Happened 16 Feb 2026Live 5 Oct 2026
ZDF broadcasts AI-generated video without labelling in heute journal report on ICE operations
On 16 February 2026, ZDF's news programme 'heute journal' broadcast a report on US border police ICE operations that included a video sequence generated by OpenAI's Sora AI, without labelling it as AI-generated. The broadcaster initially described the omission as a 'technical error' and later removed the AI material from the report, apologising for the mistake. The report also included a real video from 2022 that was used out of context.
Company
ZDF
AI system
Sora
2 source articlesheise.de
WF-EBIPSNHappened 1 Dec 2025Live 5 Oct 2026
Amazon's AI shopping tool lists products from retailers without consent
Amazon launched a program called Shop Direct and a 'Buy for Me' AI agent that scrapes products from other retailers' websites and lists them on Amazon without their permission. Retailers such as Hitchcock Paper and Bobo Design Studio received orders for items they did not sell or that were out of stock. Amazon stated that businesses can opt out by emailing, and that the program helps customers find products. Some retailers have accused Amazon of exploiting their businesses.
Company
Amazon
AI system
Shop Direct / Buy for Me
5 source articlescnbc.com
WF-6S0SOXHappened 1 Jan 2024Live 5 Oct 2026
Anthropic destroyed millions of physical books to train Claude AI model
Anthropic shredded millions of physical books under Project Panama to train its Claude AI model. Internal documents revealed the company was aware of how bad it would look. A lawsuit by authors led to a $1.5 billion settlement. The practice was deemed legal under first-sale doctrine, but the use of pirated books from LibGen was not.
Company
Anthropic
AI system
Claude
5 source articlesfuturism.com
WF-RH6QXKLive 5 Oct 2026
42,900 OpenClaw AI agents exposed, 15,200 vulnerable to RCE
SecurityScorecard's STRIKE team revealed on February 9, 2026, that approximately 42,900 OpenClaw agentic AI instances are exposed on the internet due to insecure default configurations. Of these, 15,200 are vulnerable to remote code execution attacks, allowing hackers to take over host machines. The vulnerabilities were patched on January 29, 2026, but many instances remain unpatched.
AI system
OpenClaw
5 source articleshow2shout.com
WF-N73P11Live 5 Oct 2026
AI transcription tools produce errors in social work records
AI transcription tools used by social workers in English and Scottish councils have been found to produce inaccurate transcripts and summaries, including false indications of suicidal ideation and 'gibberish' notes. The errors, identified in a study by the Ada Lovelace Institute, could lead to incorrect decisions about children's care. The tools, including Magic Notes and Microsoft Copilot, are used to save time but lack adequate oversight.
Company
Multiple local authorities in England and Scotland
AI system
Magic Notes,Microsoft Copilot
8 source articlestheguardian.com
WF-OP475CHappened 1 Jan 2018Live 5 Oct 2026
Dutch probation service's OXREC algorithm flawed, leading to incorrect recidivism risk assessments
The Dutch Inspectorate of Justice and Security (Inspectie JenV) published a report finding that the probation service's (Reclassering) OXREC algorithm contains serious flaws, including swapped formulas and incorrect numbers, causing about a quarter of risk assessments to be wrong. The algorithm, used since 2018 for about 44,000 cases per year, also uses variables that can lead to discrimination, such as neighborhood score and income. The Inspectorate recommended immediate correction or temporary suspension. The probation service announced it would temporarily stop using OXREC.
Company
Reclassering Nederland
AI system
OXREC
4 source articlesinspectie-jenv.nl
WF-GAQZA3Happened 20 Jan 2026Live 5 Oct 2026
Oxford study finds ChatGPT reproduces global inequalities
A study by the Oxford Internet Institute and the University of Kentucky found that ChatGPT systematically attributes positive characteristics to wealthier countries in the Global North and negative stereotypes to lower-income regions, particularly African countries. The research analysed 20.3 million queries and identified five types of bias, concluding that bias is a structural feature of generative AI. The study was published in the journal Platforms and Society on 20th January 2026.
Company
OpenAI
AI system
ChatGPT
4 source articlescherwell.org
WF-WTAS45Happened 13 Jan 2026Live 5 Oct 2026
Roblox's AI age verification misidentifies users' ages
Roblox launched an AI-powered age verification system that estimates users' ages from facial scans. The system has misidentified many users, placing children in adult categories and adults in teen categories, and has been tricked with avatars or altered faces. Users report loss of chat access and privacy concerns, and age-verified accounts are being sold online. Roblox acknowledged the issues and said it is working on solutions.
Company
Roblox
AI system
AI-powered age verification system
5 source articleswired.com
WF-ZDNDNDLive 5 Oct 2026
Imperial College AI stethoscope falsely identified heart failure in two-thirds of flagged patients
Researchers at Imperial College London and Imperial College Healthcare NHS Trust have developed an AI-powered stethoscope that analyses heart rhythms and performs an ECG, sending results to a doctor's smartphone. A study of more than 12,000 patients across the UK found that two-thirds of patients flagged as having heart failure did not actually have the condition. The researchers acknowledged that the false positives could cause unnecessary anxiety, and many GP surgeries stopped using or barely used the device after 12 months. The team plans to deploy the stethoscope more widely.
Company
Imperial College London and Imperial College Healthcare NHS Trust
6 source articlesfuturism.com
WF-5WFE0NLive 5 Oct 2026
Heathfield man sentenced in county's first AI-generated abusive image case
James Castell, 40, of Heathfield, was sentenced at Lewes Crown Court on 19 December after pleading guilty to possessing and distributing indecent images of children, including AI-generated pseudo-photographs. Sussex Police began investigating in October 2024 and found more than 3,800 images, along with AI image-generation software. He received an 18-month suspended sentence, a Sexual Harm Prevention Order, and was ordered to register as a sex offender.
3 source articlestheheathfieldnews.co.uk

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

Sainsbury's ejects man misidentified by facial recognition software

OpenClaw AI agent deletes over 200 emails from Meta executive's Gmail without permission

Albanian actor sues government over unauthorized use of her likeness for AI virtual minister

ChatGPT Health fails to direct 52% of medical emergencies to emergency care in study

Hollywood studios accuse ByteDance's Seedance 2.0 of mass copyright infringement

Ring's AI pet-search feature draws privacy backlash after Super Bowl ad

Thailand halts iris-scan crypto scheme and deletes 1.2m biometric records

OpenClaw vulnerabilities enable data leakage and prompt injection

Fidesz AI campaign video depicts execution of Hungarian father

Moonwell loses $1.78M after AI-generated code from Claude Opus 4.6 causes oracle pricing error

OpenAI's Sora 2 generates false videos in 80% of tests

Glasgow AI mural design sparks criticism from local artists

British Museum faces backlash over AI-generated social media images

Alpha School parents allege harm from AI-driven learning software

ZDF broadcasts AI-generated video without labelling in heute journal report on ICE operations

Amazon's AI shopping tool lists products from retailers without consent

Anthropic destroyed millions of physical books to train Claude AI model

42,900 OpenClaw AI agents exposed, 15,200 vulnerable to RCE

AI transcription tools produce errors in social work records

Dutch probation service's OXREC algorithm flawed, leading to incorrect recidivism risk assessments

Oxford study finds ChatGPT reproduces global inequalities

Roblox's AI age verification misidentifies users' ages

Imperial College AI stethoscope falsely identified heart failure in two-thirds of flagged patients

Heathfield man sentenced in county's first AI-generated abusive image case