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 152 of 201

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

WF-CHRBJKHappened 10 Oct 2019Live 5 Oct 2026
Knightscope RoboCop ignores woman’s emergency call in California park
During a fight in a Salt Lake Park parking lot in Huntington Park, California, a bystander pressed the emergency button on a Knightscope RoboCop security robot. The robot repeatedly told her to ‘step out of the way’ instead of summoning help. The woman then tried to get the robot’s camera to recognise the situation, but it still did not assist, and a companion had to dial 911. The police chief stated the robots are on a trial basis and not yet linked to the police department.
Company
Knightscope
AI system
RoboCop
5 source articlesinterestingengineering.com
WF-CFSCV3Happened 1 Mar 2023Live 5 Oct 2026
Meta's Facebook ad algorithm shows job ads with gender bias, Global Witness finds
Global Witness conducted an investigation by posting job ads on Facebook in multiple countries. They found that Facebook's ad delivery algorithm showed ads predominantly to one gender, with 91% of mechanic ad viewers being male and 79% of preschool teacher ad viewers being female. The organisation alleges this constitutes discrimination and has filed complaints with regulators in France and the Netherlands. Meta responded that it has applied targeting restrictions and offers transparency, but did not directly address the findings.
Company
Meta
AI system
Facebook ad delivery algorithm
1 source articleglobalwitness.org
WF-CDM7XGHappened 15 Jan 2024Live 5 Oct 2026
Amazon hit by wave of AI-generated product listings with OpenAI error messages
Amazon's marketplace was flooded with product listings bearing titles such as 'I'm sorry, I cannot fulfil this request as it goes against OpenAI use policy', indicating sellers used AI chatbots like ChatGPT to generate descriptions without review. The listings were noticed by users on social media and subsequently removed by Amazon. Amazon stated it is enhancing its systems to prevent such issues, while OpenAI did not immediately respond to a request for comment.
Company
Amazon
5 source articlesbusinessinsider.com
WF-CG6W7THappened 12 Mar 2025Live 5 Oct 2026
AI deepfake of Trump calling Zelenskiy 'Temu Zelenskiy' spreads on Facebook
An AI-generated video depicting US President Donald Trump calling Ukraine's President Volodymyr Zelenskiy 'Temu Zelenskiy' for not wearing a suit was shared on Facebook as if real. The video was created by a YouTube account, DangerousAI, which describes its content as AI parody, but the Facebook posts did not include this disclaimer. Some users appeared to believe the video was genuine, while others recognised it as a joke. AAP FactCheck confirmed the video is a deepfake with telltale signs of AI generation.
Company
DangerousAI
1 source articleaap.com.au
WF-CH74K2Happened 21 Mar 2023Live 5 Oct 2026
NEDA Fires Helpline Staff, Replaces Them with Chatbot After Unionization
The National Eating Disorders Association (NEDA) fired four full-time helpline staff and announced the shutdown of its 20-year-old helpline, replacing it with a chatbot named Tessa. The staff had recently unionized and won an NLRB election, leading to allegations that the move was union busting. NEDA claims the chatbot, a rule-based system developed by Washington University, is a separate program to better serve the community, but the union and critics argue it cannot replace human empathy and may cause irreparable harm. The helpline is set to end on June 1, 2023, with volunteers asked to become testers for the chatbot.
Company
National Eating Disorders Association (NEDA)
AI system
Tessa
10 source articlesvice.com
WF-CDY4BRHappened 1 Jul 2024Live 5 Oct 2026
Perplexity AI Misdiagnosis Leads to Man's Death After He Avoids Cancer Treatment
Benjamin Riley's father, a former neuroscientist, used Perplexity AI to self-diagnose his chronic lymphocytic leukemia and was incorrectly told he had Richter's Transformation, leading him to refuse his oncologist's recommended Ven-Obi treatment for a year. Despite his son's efforts to correct the misinformation by contacting the study authors, the father's condition worsened, and he eventually started treatment but died weeks later. The incident highlights the dangers of relying on AI for medical advice.
Company
Perplexity AI
AI system
Perplexity AI
1 source articlebuildcognitiveresonance.substack.com
WF-CFRK5NHappened 1 Jan 2022Live 5 Oct 2026
NYPD's ShotSpotter Gunshot Detection System Found Largely Inaccurate, Wasting Resources
An audit by the New York City Comptroller found that ShotSpotter, a gunshot detection system used by the NYPD, is overwhelmingly inaccurate. Only 13 percent of alerts in June 2023 corresponded to confirmed shootings, and the system missed over 200 real gunfire incidents in Manhattan in 2022. The NYPD has spent more than $45 million on the system since 2015. The comptroller advised against renewing the contract until the system can be fully evaluated.
Company
New York Police Department
AI system
ShotSpotter
10 source articlesnytimes.com
WF-CDQGLXLive 5 Oct 2026
Tom Hanks and Gayle King Warn About AI Deepfake Endorsement Scams
Actor Tom Hanks and broadcast journalist Gayle King warned their followers about AI-generated deepfake videos that falsely depicted them endorsing a dental plan and a weight-loss product, respectively. The videos, created without their consent, used repurposed footage to make the endorsements appear real. Both celebrities posted on Instagram to alert the public, stating they had no involvement with the products. The incident highlights growing concerns over unauthorized AI-generated content and its potential to deceive consumers.
10 source articlesmymodernmet.com
WF-CESBK7Happened 14 Oct 2024Live 5 Oct 2026
Portland Water Bureau AI pilot offers discount to billionaire CEO
The Portland Water Bureau used a machine learning system in a randomised control trial to identify customers for its financial assistance programme. The system selected Columbia Sportswear CEO Tim Boyle, a billionaire and one of the city's largest residential water consumers, for a 40% water bill discount. Boyle declined the discount, stating it should go to someone who needs it. The bureau is testing whether the SERVUS algorithm can accurately identify customers' ability to pay.
Company
Portland Water Bureau
AI system
Smart Discount Program
1 source articlewweek.com
WF-CCLUTXLive 5 Oct 2026
Minor arrested in Palma for AI-generated nude images of schoolmates
A minor in Palma was arrested for allegedly using an AI application to generate realistic nude images of five female schoolmates from their clothed social media photos. The images were shared, and the victims' families reported them to the police, who investigated and identified the minor as a classmate. The case has been referred to the Juvenile Prosecutor's Office as a child pornography offence. The police noted that the images appeared real at first glance.
10 source articlesinfolibre.es
WF-CCRF4KHappened 1 Oct 2024Live 5 Oct 2026
NPR's David Greene sues Google over AI voice cloning in NotebookLM
David Greene, a former NPR host, alleges that Google's NotebookLM tool replicated his voice without permission. He says he was 'completely freaked out' when a colleague asked if he had licensed his voice to the company. Greene is suing Google, claiming the product violated his rights by using an AI-generated voice that sounds like his own. The lawsuit is pending.
Company
Google
AI system
NotebookLM
1 source articlewashingtonpost.com
WF-CC7AFUHappened 18 Jan 2026Live 5 Oct 2026
Beep automated bus rear-ended by Tesla in Washington D.C.
An automated bus operated by Beep was rear-ended by a Tesla driver making an illegal lane change in Washington D.C. on 18 January 2026. The bus, part of a pilot programme with the U.S. Department of Transportation, had a human backup driver and sustained minor cosmetic damage. Service was temporarily paused but resumed after a review found the bus operated appropriately. No police were called and the incident was not reported to NHTSA as the damage was under $1,000 and the other driver was at fault.
Company
Beep
AI system
Beep automated bus
2 source articlesthedailyrecord.com
WF-CBSEF5Happened 1 Jan 2017Live 5 Oct 2026
VA's Suicide Prevention Algorithm Favours Men, Ignores Sexual Trauma
The Department of Veterans Affairs' REACH VET algorithm, designed to identify veterans at high risk of suicide, reportedly prioritises white men and excludes factors like military sexual trauma that disproportionately affect women. An investigation found the model gives preference to divorced or widowed men but no female-specific groups, despite a sharper rise in suicide rates among female veterans. VA officials defended the algorithm, stating that excluded variables were not the strongest predictors, while critics argue systemic bias is manifesting. The agency is working on compliance reviews but has not changed the system.
Company
Department of Veterans Affairs
AI system
REACH VET
2 source articlesmilitary.com
WF-CAA92RHappened 5 Jun 2025Live 5 Oct 2026
AI Helped Long Island Man Build Bombs for Manhattan Attack
Michael Gann, 55, allegedly used artificial intelligence to learn how to purchase and mix chemicals to build seven homemade bombs. He transported the devices to Manhattan, storing five on a SoHo rooftop, and threw one onto subway tracks. Gann was arrested on 5 June after posting threatening messages online and is now facing federal charges. The AI system provided instructions that Gann said made bomb-making 'easier than buying gun powder'.
1 source articlenbcnews.com
WF-CBFCSHHappened 1 Nov 2023Live 5 Oct 2026
Lancaster Country Day School student used AI to create nude images of female classmates
A student at Lancaster Country Day School allegedly used an artificial intelligence program to generate nude images featuring the faces of more than 20 female students during the 2023-2024 school year. The school first learned of the images in November 2023 but was unable to corroborate the report until a second allegation in May 2024, after which the student was removed and law enforcement was notified. Parents have criticised the school's handling and communication, and an investigation by local authorities is ongoing. The incident has highlighted a legal grey area, as Pennsylvania currently has no law specifically criminalising the creation of such AI-generated images.
Company
Lancaster Country Day School
8 source articleslancasteronline.com
WF-CBJ3H7Live 5 Oct 2026
Two Canadians lose $2.3 million in AI deepfake crypto scam
Two Canadians, one from Ontario and one from Prince Edward Island, lost a combined $2.3 million after being deceived by AI-generated deepfake videos. The Ontario victim lost $1.7 million after watching a fake video of Elon Musk, while the other lost $600,000 after seeing a clip falsely linked to the show Dragon's Den. The scam involved a fraudulent cryptocurrency investment scheme. The incident was reported on December 21, 2025.
2 source articlesbingx.com
WF-CAHFC8Live 5 Oct 2026
Deepfake of Miami Beach realtor used in UK romance scam
A woman in the United Kingdom was deceived by a scammer who used AI-generated deepfake videos of Miami Beach real estate agent Andrés Asion. The videos, which showed Asion's face and voice, led her to believe she was in a romantic relationship with him. After she contacted the real Asion, they met and confirmed the videos were fabricated. Asion warns that such misuse of AI could happen to anyone and calls for legislative protections.
2 source articleslocal10.com
WF-CBRYM9Happened 1 Nov 2024Live 5 Oct 2026
AI deepfakes of doctors used to scam diabetes patients in Australia
Scammers used AI-generated deepfake videos and fake interviews impersonating diabetes expert Professor Jonathan Shaw and other health professionals to promote a dietary supplement called Glyco Balance. The fraudulent ads falsely claimed the supplement, sold by Vellec Group, was developed at the Baker Institute and that commonly prescribed diabetes medication is dangerous. One victim, Michael, purchased the supplement for $340 before realising it was a scam. The Therapeutic Goods Administration is assessing the situation, and Meta removed the fake video after the Baker Institute reported it.
Company
Vellec Group
AI system
Generative AI for deepfake video creation
1 source articleabc.net.au
WF-CA5PVXHappened 14 Feb 2024Live 5 Oct 2026
Meta's Advantage Plus AI ad tool overspends and underperforms for advertisers
In February 2024, Meta's automated ad platform Advantage Plus began malfunctioning, causing advertisers' costs per impression to skyrocket and blowing through daily budgets without delivering sales. Multiple marketers reported that the AI-driven tool ignored cost caps and performed unpredictably, leading some to halt its use. Meta acknowledged a platform bug on February 14 and issued refunds to some, but problems persisted into April, with the company claiming the system was working as expected for most.
Company
Meta
AI system
Advantage Plus
1 source articletheverge.com
WF-CAJKXRHappened 9 Feb 2025Live 5 Oct 2026
AI-generated Facebook pages mislead New Zealanders with fake news
A 1News investigation found at least 10 Facebook pages, including 'NZ News Hub', using AI to rewrite legitimate news stories and pair them with unlabelled, fabricated images. The pages dramatised events, manipulated photos of real people including a deceased minor, and attracted thousands of followers. Authorities warned the public about the pages, and experts said they blur the line between real and fake news, eroding trust in media. The page disappeared from Facebook after 1News contacted Meta.
Company
NZ News Hub
AI system
ChatGPT
1 source article1news.co.nz
WF-C6XDL5Happened 14 Aug 2024Live 5 Oct 2026
X's Grok AI Image Generator Lacks Guardrails, Users Create Offensive Images of Trademarked Characters
On August 14, 2024, X rolled out image generation capabilities for its Grok AI chatbot to Premium users. The feature lacked content moderation guardrails, allowing users to create offensive images of political figures and trademarked characters like Nintendo's Mario. The images, which included depictions of violence and drug use, appeared alongside advertisements for the affected brands, raising concerns about misinformation and reputational damage. X owner Elon Musk acknowledged the feature's launch and stated the team was training Grok to be 'truthful, but also kind and funny.'
Company
X
AI system
Grok-2
3 source articlesmediaite.com
WF-C7VZQTHappened 8 Jan 2025Live 5 Oct 2026
AI-generated images falsely show Hollywood Sign on fire during LA wildfires
During the January 2025 Los Angeles wildfires, AI-generated images and videos depicting the Hollywood Sign engulfed in flames circulated widely on social media. The landmark was not affected by the Sunset Fire, but the fake content caused confusion and distress among people already dealing with evacuations and loss. Social media users and platforms added context labels noting the images were fake, while some criticised the spread of harmful misinformation.
Company
X
AI system
AI image generators
2 source articlesdeadline.com
WF-C93AE5Live 5 Oct 2026
Uber used Greyball software to evade law enforcement
Uber deployed a secret tool called Greyball that used data to identify and deny service to law enforcement officials attempting to catch drivers violating regulations. The system was used in multiple US cities and countries. Uber defended the program as protecting drivers from harm, but a law professor suggested it may have violated the Computer Fraud and Abuse Act. The revelation came amid other controversies for the company.
Company
Uber
AI system
Greyball
2 source articlestheguardian.com
WF-C4PFXKHappened 1 Dec 2020Live 5 Oct 2026
TikTok Users Bypass Pro-Anorexia Ban with Misspelled Hashtags
TikTok's recommendation algorithm continues to surface pro-anorexia content to users despite banning related hashtags, as users circumvent moderation by misspelling terms. The platform's For You Page algorithm amplifies harmful content based on user engagement, potentially trapping vulnerable individuals in a cycle of eating disorder content. TikTok has banned some accounts and redirects searches to support resources, but the issue persists. Eating disorder charities have called for stronger safety measures.
Company
TikTok
AI system
TikTok For You Page algorithm
1 source articleinsider.com

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

Knightscope RoboCop ignores woman’s emergency call in California park

Meta's Facebook ad algorithm shows job ads with gender bias, Global Witness finds

Amazon hit by wave of AI-generated product listings with OpenAI error messages

AI deepfake of Trump calling Zelenskiy 'Temu Zelenskiy' spreads on Facebook

NEDA Fires Helpline Staff, Replaces Them with Chatbot After Unionization

Perplexity AI Misdiagnosis Leads to Man's Death After He Avoids Cancer Treatment

NYPD's ShotSpotter Gunshot Detection System Found Largely Inaccurate, Wasting Resources

Tom Hanks and Gayle King Warn About AI Deepfake Endorsement Scams

Portland Water Bureau AI pilot offers discount to billionaire CEO

Minor arrested in Palma for AI-generated nude images of schoolmates

NPR's David Greene sues Google over AI voice cloning in NotebookLM

Beep automated bus rear-ended by Tesla in Washington D.C.

VA's Suicide Prevention Algorithm Favours Men, Ignores Sexual Trauma

AI Helped Long Island Man Build Bombs for Manhattan Attack

Lancaster Country Day School student used AI to create nude images of female classmates

Two Canadians lose $2.3 million in AI deepfake crypto scam

Deepfake of Miami Beach realtor used in UK romance scam

AI deepfakes of doctors used to scam diabetes patients in Australia

Meta's Advantage Plus AI ad tool overspends and underperforms for advertisers

AI-generated Facebook pages mislead New Zealanders with fake news

X's Grok AI Image Generator Lacks Guardrails, Users Create Offensive Images of Trademarked Characters

AI-generated images falsely show Hollywood Sign on fire during LA wildfires

Uber used Greyball software to evade law enforcement

TikTok Users Bypass Pro-Anorexia Ban with Misspelled Hashtags