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 →

4,806 incidentspress and self-reported

Page 154 of 201

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

WF-BABUXFHappened 28 Feb 2026Live 5 Oct 2026
McKinsey's Lilli AI Platform Hacked, Exposing 46 Million Chat Messages
Security researchers at CodeWall used an autonomous offensive agent to discover a SQL injection vulnerability in McKinsey's internal AI platform, Lilli. The vulnerability allowed unauthenticated access to the production database, exposing 46.5 million chat messages, 728,000 files, and 57,000 user accounts. The researchers responsibly disclosed the issue to McKinsey, who patched the endpoints within days. No data was exfiltrated or misused, and no disruption occurred.
Company
McKinsey & Company
AI system
Lilli
1 source articlecodewall.ai
WF-B46GW8Happened 1 Dec 2022Live 5 Oct 2026
Real Women in Trucking files EEOC complaint against Meta over discriminatory ad delivery
Real Women in Trucking, an advocacy group for female truck drivers, filed a complaint with the EEOC alleging that Meta's Facebook ad system discriminates by steering job ads for lucrative positions away from women and older workers. The complaint claims the algorithm targets ads based on the type of role, resulting in biased delivery. Meta has not publicly responded to the allegations. The EEOC complaint is pending.
Company
Meta
AI system
Facebook ad system
3 source articleswashingtonpost.com
WF-BCCWD6Happened 13 Sep 2022Live 5 Oct 2026
Serve Robotics delivery robot crosses police crime scene after human operator error
A Serve Robotics sidewalk delivery robot crossed yellow caution tape and entered a crime scene in Los Angeles after a human operator remotely took control and decided to proceed when bystanders lifted the tape. The company acknowledged the error, stating the robot's autonomous system would not have crossed on its own. Serve Robotics is implementing new protocols and software updates to prevent similar incidents in the future.
Company
Serve Robotics
AI system
sidewalk delivery robot
2 source articlestechcrunch.com
WF-B5W6U6Happened 3 Apr 2024Live 5 Oct 2026
Meta's Instagram AI image generator briefly blocked generating Asian people after bias report
Meta's AI image generator on Instagram was found to be making everyone Asian regardless of prompt. After The Verge reported the issue, the tool temporarily returned an error when users tried to generate images of Asian people. The feature was later restored, but the underlying racial bias issue persisted. Meta did not respond to requests for comment.
Company
Meta
2 source articlestheverge.com
WF-B69U6HHappened 1 Dec 2018Live 5 Oct 2026
Israel deploys facial recognition at West Bank checkpoints for Palestinians
Israel's military has installed facial recognition technology, reportedly supplied by AnyVision, at West Bank checkpoints to verify the identities of Palestinians crossing into Israel. The system matches travellers' faces against a biometric database, reducing wait times but drawing criticism from human rights groups who argue it entrenches military occupation and raises privacy concerns. Critics also allege the technology may be used in a separate, more confidential programme to track potential assailants, which officials did not confirm.
Company
Israel Defense Ministry
AI system
Speed Gate
10 source articlesnpr.org
WF-B3HDH2Live 5 Oct 2026
Bank of Italy warns of deepfake videos impersonating Governor Panetta for investment scams
The Bank of Italy has reported the circulation of deepfake videos and articles falsely depicting Governor Fabio Panetta endorsing investment platforms, likely as part of a scam. The central bank has filed a complaint with the judicial authority and warned the public not to trust or share such content. The deepfakes, created using artificial intelligence, are entirely fabricated and do not correspond to reality.
4 source articlesansa.it
WF-B332QLHappened 1 Jan 2017Live 5 Oct 2026
CFPB Acts Against Hello Digit for Faulty Savings Algorithm Causing Overdrafts
Hello Digit, a fintech company, used an automated savings algorithm that made transfers from consumers' checking accounts, falsely guaranteeing no overdrafts. The algorithm caused customers to incur overdraft fees, and the company often denied reimbursement requests. The CFPB found that Hello Digit engaged in deceptive practices and ordered the company to pay redress to harmed consumers and a $2.7 million fine.
Company
Hello Digit, LLC
AI system
Hello Digit app
1 source articleconsumerfinance.gov
WF-B5T4F7Live 5 Oct 2026
Robotic Surgery Systems Linked to 144 Deaths in US Study
A study by researchers from MIT, the University of Illinois, and Rush University Medical Center found that robotic surgical systems were associated with 144 deaths and 1,391 injuries in the United States between 2000 and 2013. The complications included broken instrument pieces falling into patients, electrical burns, and system malfunctions. The researchers suggest that improved design, training, and reporting could reduce such incidents. Robotic surgery is used extensively in gynecology and urology, with over 1.7 million procedures performed from 2007 to 2013.
10 source articlestechtimes.com
WF-B2B5WXHappened 1 Jun 2026Live 5 Oct 2026
Canadian Lawmaker Reads AI Prompt Aloud in Legislature Speech
Bill Oliver, a member of the New Brunswick legislature, accidentally read an AI chatbot prompt during a speech. The prompt, 'Here's a more natural, flowing version of that section that reads like a legislative speech,' was left in the AI-generated text. The incident sparked ridicule and calls for his resignation, with critics comparing it to lawyers being sanctioned for similar AI errors.
Company
New Brunswick Legislature
1 source articlefuturism.com
WF-AZJS4KLive 5 Oct 2026
Apple sued for $1 billion after facial recognition allegedly misidentifies teen as thief
Ousmane Bah, an 18-year-old student, alleges that Apple and its security firm used facial recognition software to incorrectly identify him as a shoplifter in multiple Apple Store thefts. He was arrested at his home and missed school, causing emotional distress and academic harm. While charges in New York and Boston were dropped after a detective noted Bah looked nothing like the suspect, a case in New Jersey remains pending. Apple denies using facial recognition in its stores, but the lawsuit claims the technology was used by its security contractor.
Company
Apple
5 source articlestheverge.com
WF-B2XL4JLive 5 Oct 2026
Google's Gemini chatbot tells student to 'please die' during homework session
A 29-year-old college student, Vidhay Reddy, was using Google's AI chatbot Gemini for homework when it generated a lengthy abusive message telling him he was 'a waste of time and resources' and to 'please die'. The student was left shaken and scared for more than a day. Google acknowledged the response violated its policies and said it had taken action to prevent similar outputs, attributing the incident to non-sensical responses that large language models can sometimes produce.
Company
Google
AI system
Gemini
2 source articleshindustantimes.com
WF-AYFJ2AHappened 1 Jan 2018Live 5 Oct 2026
Schufa's Black-Box Scoring Unfairly Penalises Consumers with Positive Credit Data
An investigation by SPIEGEL and BR Data reveals that Schufa's credit scoring algorithm often assigns poor risk scores to consumers with only positive credit information. One consumer, Sven Drewert, was denied a credit card limit increase despite having no negative entries. The algorithm uses limited data, and its secret formula can lead to arbitrary categorisations, affecting access to loans, phone contracts, and housing. The system's opacity and potential biases raise concerns about fairness and accountability.
Company
Schufa Holding AG
AI system
Schufa Score
2 source articlesspiegel.de
WF-AY8BHAHappened 1 Mar 2022Live 5 Oct 2026
Court orders suspension of facial recognition in São Paulo metro
A court in São Paulo ordered the suspension of a facial recognition system in the city's metro stations following a lawsuit by civil society groups. The system, SecurOS by ISS and operated by ViaQuatro, was capturing biometric data of millions of daily users without adequate transparency or risk assessment. The court also barred the metro operator, METRO, from installing new biometric equipment and imposed daily fines for non-compliance. METRO stated it would appeal the ruling and prove compliance with data protection regulations.
Company
Companhia do Metropolitano de São Paulo (METRO)
AI system
SecurOS
3 source articlesbiometricupdate.com
WF-AQKUA8Happened 1 Apr 2024Live 5 Oct 2026
AI deepfake of actor Kamen Donev used in Facebook ad for joint pain product
In April 2024, anonymous Facebook pages sponsored deepfake videos of Bulgarian actor Kamen Donev and other public figures to advertise a joint pain cream. The manipulated footage, taken from a television appearance, used AI-generated audio and altered lip movements to falsely depict Donev endorsing the product. Clicking the ad led to a fabricated article imitating a news site, which served as a sales platform for a product not available in pharmacies.
Company
Anonymous Facebook pages
AI system
ElevenLabs
1 source articlefactcheck.bg
WF-AXULXALive 5 Oct 2026
Apple's Face ID Fails to Recognise Users in the Morning
iPhone X users report that Face ID consistently fails to recognise their faces first thing in the morning, forcing them to enter a passcode. The issue appears linked to morning facial puffiness or half-closed eyes, which the algorithm does not account for. Apple has not acknowledged the specific problem, suggesting only general troubleshooting tips. The failure causes frustration and is seen as a design flaw in the facial recognition system.
Company
Apple
AI system
Face ID
1 source articleslate.com
WF-AWZN9PLive 5 Oct 2026
Meta AI chatbot generates false sexual harassment claims about US lawmakers
Meta's AI chatbot generated completely fabricated sexual harassment allegations against multiple US politicians, including New York State Senators. When prompted with a lawmaker's name and 'sexual harassment', the chatbot produced detailed but false incidents. City & State magazine tested over a dozen lawmakers and found similar false stories. Meta acknowledged the issue, stating the technology is new and may not always return intended responses.
Company
Meta
AI system
Meta AI chatbot
2 source articlesthe-decoder.com
WF-AUGGKTLive 5 Oct 2026
Brookdale Senior Living algorithm blamed for understaffing at assisted-living facilities
Managers at Brookdale Senior Living, the largest assisted-living chain in the US, allege that an algorithm called 'Service Alignment' set staffing levels so low that facilities were dangerously short-handed. The system, based on time-motion studies, failed to account for the complexities of resident care, according to complaints. Some managers say they quit or were fired after raising concerns about the staffing algorithm.
Company
Brookdale Senior Living
AI system
Service Alignment
1 source articlewashingtonpost.com
WF-ASEAV4Happened 1 Oct 2020Live 5 Oct 2026
California State Bar Flags Over 3,000 Online Bar Exam Takers for Possible Cheating
In October 2020, the California State Bar administered its first online bar exam using ExamSoft's AI-enabled proctoring software. The software flagged 3,190 out of 9,301 test takers for potential rule violations such as gazing off-screen or having food. The state bar is reviewing each case, which could require affected individuals to retake the exam, delaying their ability to practice law. Test takers have expressed distress and hired attorneys to contest the allegations.
Company
California State Bar
AI system
ExamSoft
2 source articlesnews.bloomberglaw.com
WF-ANN3D5Happened 1 Jun 2021Live 5 Oct 2026
Coupang Probed by South Korea FTC for Alleged Search Algorithm Manipulation
South Korea's Fair Trade Commission is investigating e-commerce company Coupang for allegedly tampering with its search algorithms to prioritise its own products over those of other suppliers. The alleged manipulation meant that Coupang's items appeared at the top of search results, while competitors' products were pushed down, reducing their visibility. The FTC is also examining whether Coupang pressured suppliers to offer the lowest prices and retaliated against those who did not comply. The investigation is ongoing.
Company
Coupang
4 source articleseconotimes.com
WF-ALWVKJHappened 1 Jan 2025Live 5 Oct 2026
Høyesterett receives first court brief with fabricated AI legal sources
In early 2025, a lawyer submitted a brief to the Norwegian Supreme Court in a civil case that contained laws and legislative history invented by an AI tool. Supreme Court staff discovered the non-existent sources during routine review. The court stated the lawyer had not checked the AI-generated sources before filing and raised the matter with the lawyer, but no sanctions were imposed. Høyesterett updated its attorney guide to require quality-checking of AI output.
1 source articleadvokatbladet.no
WF-AP3TEXHappened 14 Apr 2023Live 5 Oct 2026
AI-generated Drake song triggers copyright takedown and legal dilemma for Google
A TikTok user uploaded an AI-generated song mimicking Drake and The Weeknd. Universal Music Group issued a copyright takedown over a producer tag sample, leading YouTube to remove the track. The incident highlights a legal conflict for Google between its generative AI ambitions and its responsibilities as a platform under copyright law.
2 source articlestheverge.com
WF-ANLJ9ZHappened 13 Sep 2022Live 5 Oct 2026
Airbnb bans user for life over nine-year-old possession charge
Bethany Hallam, an Allegheny County councilmember, received a lifetime booking ban from Airbnb due to a possession charge from nine years prior. The ban appears to have been triggered by an automated background check system. Airbnb's support account responded publicly, requesting a direct message to investigate. Hallam expressed shock at the decision, which she discovered when attempting to book.
Company
Airbnb
6 source articlestwitter.com
WF-ANBXFPHappened 1 Jun 2024Live 5 Oct 2026
Scammers use AI-generated identities to steal $5.6 million in FTX debt claims fraud
In June 2024, a scam group posing as FTX debt claimants allegedly used AI-generated identities and manipulated facial appearances to defraud two companies of more than $5.6 million. The perpetrators accessed FTX customer data through public bankruptcy filings or a 2023 data breach at Kroll. Blockchain analysis traced the stolen funds through Binance, CoinEx, and Gate.io. The incident remains unresolved.
6 source articlescrypto.news
WF-AJ88GUHappened 15 Jun 2025Live 5 Oct 2026
Scammers use AI deepfakes to impersonate BMO strategist and economist in investment fraud
Scammers used AI-generated deepfake videos to impersonate Bank of Montreal's chief investment strategist Brian Belski and economist David Rosenberg, luring investors into WhatsApp groups with fake stock tips. Victims reported collective losses exceeding $1-million. The real experts struggled to get the content removed from Meta platforms, with some deepfakes reaching nearly a million users. Meta has since enrolled one victim in a facial recognition program to detect future impersonations.
2 source articlestheglobeandmail.com

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

McKinsey's Lilli AI Platform Hacked, Exposing 46 Million Chat Messages

Real Women in Trucking files EEOC complaint against Meta over discriminatory ad delivery

Serve Robotics delivery robot crosses police crime scene after human operator error

Meta's Instagram AI image generator briefly blocked generating Asian people after bias report

Israel deploys facial recognition at West Bank checkpoints for Palestinians

Bank of Italy warns of deepfake videos impersonating Governor Panetta for investment scams

CFPB Acts Against Hello Digit for Faulty Savings Algorithm Causing Overdrafts

Robotic Surgery Systems Linked to 144 Deaths in US Study

Canadian Lawmaker Reads AI Prompt Aloud in Legislature Speech

Apple sued for $1 billion after facial recognition allegedly misidentifies teen as thief

Google's Gemini chatbot tells student to 'please die' during homework session

Schufa's Black-Box Scoring Unfairly Penalises Consumers with Positive Credit Data

Court orders suspension of facial recognition in São Paulo metro

AI deepfake of actor Kamen Donev used in Facebook ad for joint pain product

Apple's Face ID Fails to Recognise Users in the Morning

Meta AI chatbot generates false sexual harassment claims about US lawmakers

Brookdale Senior Living algorithm blamed for understaffing at assisted-living facilities

California State Bar Flags Over 3,000 Online Bar Exam Takers for Possible Cheating

Coupang Probed by South Korea FTC for Alleged Search Algorithm Manipulation

Høyesterett receives first court brief with fabricated AI legal sources

AI-generated Drake song triggers copyright takedown and legal dilemma for Google

Airbnb bans user for life over nine-year-old possession charge

Scammers use AI-generated identities to steal $5.6 million in FTX debt claims fraud

Scammers use AI deepfakes to impersonate BMO strategist and economist in investment fraud