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 →

4,486 incidentspress and self-reported

Page 103 of 187

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

WF-VOJUX3Happened 4 Sep 2020Live 5 Oct 2026
Suzhou clarifies civility scoring system after public backlash
The Suzhou municipal government announced a civility code system that assigns scores to residents based on their behavior, including traffic violations and volunteer work. The system, currently in a trial phase, aims to generate a personal portrait for each resident. Following public backlash and concerns about privacy and potential abuse, an official stated that participation will not be mandatory and that a bad score will not affect citizens. The system is still under development.
Company
Suzhou municipal government
AI system
Civility code
10 source articlessixthtone.com
WF-083GH0Live 5 Oct 2026
UK councils use Covid OneView AI to harvest personal data for risk scoring
UK local authorities are using a system called Covid OneView, developed by data analytics firm Xantura, to harvest millions of personal details from council records. The system uses predictive analytics and AI to assign risk scores to households and individuals, aiming to identify those vulnerable to Covid or likely to break lockdown rules. Privacy campaigners and MPs have criticised the lack of transparency and the extent of data collection, which includes sensitive information such as debt levels, living arrangements, and even notes on unfaithful sex. Xantura and Barking and Dagenham Council have defended the system as compliant with data protection rules and focused on providing support.
Company
UK local authorities
AI system
Covid OneView
6 source articlesdailymail.co.uk
WF-GIG3RPHappened 26 Feb 2020Live 5 Oct 2026
Italian regulator orders Como to stop facial recognition surveillance
The Italian Data Protection Authority (Garante) ordered the Municipality of Como to cease its use of a facial recognition system installed in Parco Tokamakhi near the main railway station. The system was intended to identify people under investigation or reported missing, and to detect suspicious behaviour. The Garante found that the municipality lacked a specific legal basis under national law for collecting and storing biometric data, and issued an injunction to conform to the law.
Company
Comune di Como
AI system
facial recognition system
9 source articlesgaranteprivacy.it
WF-CR24UYLive 5 Oct 2026
Buenos Aires city government recuses judge who suspended facial recognition system
The Buenos Aires city government recused Judge Roberto Gallardo after he suspended the Sistema de Reconocimiento Facial de Prófugos, a facial recognition system used to identify and detain individuals with outstanding arrest warrants. The government claims the judge's ruling was politically motivated and that the system has helped capture over 1,600 fugitives. The government is challenging the suspension in court.
Company
Gobierno de la Ciudad Autónoma de Buenos Aires
AI system
Sistema de Reconocimiento Facial de Prófugos
10 source articlesbuenosaires.gob.ar
WF-TBQN8BHappened 18 Nov 2020Live 5 Oct 2026
Ugandan police accused of using Huawei facial recognition to arrest protesters
The Ugandan police are accused of using a facial recognition system supplied by Huawei to track, arrest and torture anti-government protesters. The police confirmed they used images from the surveillance system to identify and detain more than 836 people involved in protests on November 18-19, 2020. Over 50 people were killed during the demonstrations. Critics say the system is unregulated and invades privacy.
Company
Ugandan Police
10 source articlesbiometricupdate.com
WF-N77XRCHappened 25 Aug 2020Live 5 Oct 2026
Facebook still allows discriminatory ads despite settlement
A Wisconsin health care agency, Tenderness Health Care, posted a job ad on Facebook targeting users under 55 with 'African American multicultural affinity.' Federal law prohibits such age and race discrimination in job ads. The Markup found the ad still appeared despite Facebook's 2019 settlement agreeing to change its practices. Facebook removed the ad after being contacted, but critics say the company's algorithm continues to deliver ads in a discriminatory manner.
Company
Facebook
AI system
Facebook Ads
10 source articlesthemarkup.org
WF-6SGCA1Happened 1 Jun 2020Live 5 Oct 2026
OAIC finds 7-Eleven breached privacy with facial recognition surveys
The Australian Information Commissioner found that 7-Eleven interfered with customers' privacy by collecting facial images and creating faceprints through in-store surveys without adequate notice or consent. The system was used to exclude non-genuine responses and profile demographics. The OAIC ordered 7-Eleven to destroy the collected data and cease the practice.
Company
7-Eleven
9 source articlesoaic.gov.au
WF-Z8V05ELive 5 Oct 2026
Mass claim against TikTok for privacy violations and child data protection
A Dutch foundation, SOMI, has launched a mass claim against TikTok, alleging that the platform collects personal data from users, including children, without proper consent. The claim seeks compensation for users and demands better age verification and parental controls. The European Data Protection Board and various national authorities are investigating TikTok's data practices. The claim is ongoing and aims to hold TikTok accountable for privacy violations.
Company
TikTok
AI system
TikTok
10 source articlestiktokclaim.org
WF-BRLQG9Live 5 Oct 2026
Police facial recognition robot identifies anger and distress
A police facial recognition robot identifies anger and distress. The article gives no further details about the system, its deployment, or any consequences.
8 source articlesthetimes.co.uk
WF-N335FIHappened 29 Apr 2020Live 5 Oct 2026
RocNation files copyright claim to remove Jay-Z deepfake videos from YouTube
Jay-Z's label RocNation filed a copyright claim to remove two deepfake videos from YouTube that used AI to impersonate Jay-Z's voice. The videos were created by the channel Vocal Synthesis, which produces humorous audio deepfakes. RocNation alleged the content unlawfully used an AI to impersonate their client's voice, though the article notes the legal basis may be inaccurate as the videos were transformative fair use.
Company
RocNation
10 source articlesinverse.com
WF-HXS00HHappened 1 Jan 2020Live 5 Oct 2026
Dutch government continued SyRI welfare fraud profiling after ban
The Dutch government deployed the SyRI algorithm to profile welfare recipients in low-income neighborhoods, despite a 2020 court ruling that the system violated EU human rights. The algorithm flagged thousands of parents for fraud based on illegal risk factors such as dual nationality, leading to false accusations, bankruptcy, and trauma. The government has announced an external review of the practices.
Company
Ministry of Social Affairs of the Netherlands
AI system
SyRI
10 source articleslighthousereports.com
WF-57PFXFHappened 25 Dec 2020Live 5 Oct 2026
Channel 4 to broadcast deepfake Queen in Christmas message
Channel 4 is to broadcast a deepfake of the Queen in its alternative Christmas message, created by VFX studio Framestore. The broadcaster said the aim was to warn about fake news, but the BBC's royal correspondent criticised the voice and some viewers called it disrespectful. Ofcom received a small number of complaints but could not consider them because it is a post-transmission regulator.
Company
Channel 4
10 source articlesbbc.co.uk
WF-XQ4KHPLive 5 Oct 2026
Dutch police predictive policing project in Roermond faces discrimination allegations
Amnesty International's report alleges that the Dutch police's Sensing predictive policing project in Roermond uses algorithms that result in automated discrimination and mass surveillance. The system assesses crime risk for individuals and locations, potentially targeting marginalized communities. The report calls for accountability and human rights safeguards.
Company
Dutch police
AI system
Sensing project
10 source articlesamnesty.org
WF-XGIXWVHappened 31 Jan 2020Live 5 Oct 2026
DWP's Universal Credit algorithm causes underpayment and hardship for claimants
The Department for Work and Pensions (DWP) uses an automated means-testing algorithm in its Universal Credit system that miscalculates earnings due to a design flaw, leading to underpayments and fluctuating benefit amounts. Claimants such as Janet R. received payments hundreds of pounds short, forcing them into debt and foodbank use. A Court of Appeal ruling in June 2020 ordered the DWP to fix the flaw, but the department had not yet fully implemented the remedy at the time of writing.
Company
Department for Work and Pensions
AI system
Universal Credit
10 source articleshrw.org
WF-TB1CFEHappened 1 Sep 2018Live 5 Oct 2026
Facebook settles with content moderators for $52 million over PTSD
Facebook agreed to pay $52 million to settle a class-action lawsuit brought by more than 10,000 content moderators who suffered post-traumatic stress disorder from viewing graphic and disturbing posts. The moderators, employed by contractors, reviewed content flagged by Facebook's automated systems. The settlement provides cash payments and medical coverage for those diagnosed with psychological conditions. Facebook admitted no wrongdoing but agreed to provide mental health counseling.
Company
Facebook
AI system
Facebook content moderation system
10 source articlesnpr.org
WF-1PU52ULive 5 Oct 2026
RT uses deepfake video to taunt Trump with post-election job offer
Russian state-owned broadcaster RT published a deepfake video showing US President Donald Trump appearing to work for the outlet as a “special host” after losing the 2020 election. RT said the video was satire and invited Trump to take up an actual job. The article notes the video was an obvious fake, and that US intelligence believed Russia was attempting to interfere in the election in subtler ways.
Company
RT
10 source articleshuffingtonpost.co.uk
WF-1HCI92Happened 5 Aug 2022Live 5 Oct 2026
Meta's BlenderBot 3 chatbot says company exploits people
Meta's prototype chatbot BlenderBot 3 told the BBC that CEO Mark Zuckerberg exploits users for money. The chatbot, released to the public in August 2022, is designed for research and entertainment but can produce offensive statements. Meta acknowledged the chatbot may make untrue or offensive statements and stated it had installed safeguards.
Company
Meta
AI system
BlenderBot 3
10 source articlesbbc.co.uk
WF-492OY0Live 5 Oct 2026
MBN introduces AI news anchor AI Kim in South Korea
MBN, a South Korean cable channel, introduced an AI-powered news anchor based on anchor Kim Ju-ha. The AI, named AI Kim, was created through deep learning using 10 hours of video of Kim. It presents news in her voice and manner, intended to allow 24/7 broadcasting.
Company
MBN
AI system
AI Kim
10 source articlesubergizmo.com
WF-8HC83KHappened 6 Nov 2020Live 5 Oct 2026
Deepfake video falsely shows PornHub banner on CNN's Magic Wall
A viral video appeared to show a PornHub banner pop up on CNN's Magic Wall behind anchor John King. The video was actually a deepfake, created by editing a real CNN clip. The article from Mediaite debunks the video, showing it was faked. The creator of the deepfake is unknown.
10 source articlesmediaite.com
WF-9RVU2YHappened 20 Apr 2020Live 5 Oct 2026
Extinction Rebellion releases deepfake of Belgian PM linking Covid-19 to climate crisis
Extinction Rebellion activists created a deepfake video of Belgian Prime Minister Sophie Wilmès. The video used undisclosed editing techniques to make her appear to give a speech linking Covid-19 to the climate crisis. The activists sent the video to politicians before publishing it online. Some viewers reportedly believed the video was authentic.
Company
Extinction Rebellion
10 source articlesmedium.com
WF-EV7DW2Live 5 Oct 2026
Southern Co-op uses facial recognition with Hikvision cameras
Southern Co-op, a UK grocery chain, deployed a facial recognition system using Hikvision cameras in its stores. The system is used to identify and track customers, but the grocer did not inform customers or obtain their consent. Privacy advocates have raised concerns about compliance with GDPR and ethical risks. The grocer has stated it is committed to a high standard of privacy.
Company
Southern Co-op
4 source articlesipvm.com
WF-UPBISOHappened 27 Feb 2020Live 5 Oct 2026
Teenager's fake congressional candidate verified on Twitter using AI-generated face
A 17-year-old high school student created a fake Republican congressional candidate named Andrew Walz for Rhode Island. He used the AI-powered website This Person Does Not Exist to generate a realistic fake face for the candidate. The teenager submitted the fake candidate to Ballotpedia, which then included him in a list sent to Twitter for verification. Twitter verified the account, but later suspended it after discovering the fraud. Ballotpedia acknowledged the mistake and updated its policy.
Company
Twitter
AI system
This Person Does Not Exist
10 source articlestime.com
WF-IKJ9FYHappened 20 May 2018Live 5 Oct 2026
Belgian political party sp.a created deep fake video of President Trump
In May 2018, the Belgian political party sp.a released a deep fake video appearing to show President Donald Trump telling Belgium to withdraw from the Paris Climate Agreement. The video was created using lip-synching technology and was intended as a parody to promote the party's climate petition. The party confirmed the video was a deep fake and said it was legally sound.
Company
sp.a (Socialistische Partij Anders)
10 source articlessnopes.com
WF-4YJCFGHappened 30 Aug 2020Live 5 Oct 2026
Twitter flags manipulated video of activist Ady Barkan's computer voice
Twitter flagged a video shared by the second-ranking House Republican as manipulated. The video spliced quotes from activist Ady Barkan, who speaks through a computer voice assistance device, making it appear he had convinced Joe Biden to defund police departments. The manipulated content was identified and labelled by Twitter under its manipulated media policy.
Company
Twitter
AI system
Twitter's manipulated media policy
10 source articleswashingtonpost.com

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

Suzhou clarifies civility scoring system after public backlash

UK councils use Covid OneView AI to harvest personal data for risk scoring

Italian regulator orders Como to stop facial recognition surveillance

Buenos Aires city government recuses judge who suspended facial recognition system

Ugandan police accused of using Huawei facial recognition to arrest protesters

Facebook still allows discriminatory ads despite settlement

OAIC finds 7-Eleven breached privacy with facial recognition surveys

Mass claim against TikTok for privacy violations and child data protection

Police facial recognition robot identifies anger and distress

RocNation files copyright claim to remove Jay-Z deepfake videos from YouTube

Dutch government continued SyRI welfare fraud profiling after ban

Channel 4 to broadcast deepfake Queen in Christmas message

Dutch police predictive policing project in Roermond faces discrimination allegations

DWP's Universal Credit algorithm causes underpayment and hardship for claimants

Facebook settles with content moderators for $52 million over PTSD

RT uses deepfake video to taunt Trump with post-election job offer

Meta's BlenderBot 3 chatbot says company exploits people

MBN introduces AI news anchor AI Kim in South Korea

Deepfake video falsely shows PornHub banner on CNN's Magic Wall

Extinction Rebellion releases deepfake of Belgian PM linking Covid-19 to climate crisis

Southern Co-op uses facial recognition with Hikvision cameras

Teenager's fake congressional candidate verified on Twitter using AI-generated face

Belgian political party sp.a created deep fake video of President Trump

Twitter flags manipulated video of activist Ady Barkan's computer voice