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.
Press reportingSelf-reported: a person’s own public post · 10 creditsSelf-reported: told to us directly · 25 credits
An anonymous Google whistleblower states that the company acquired medical data of 50 million patients from Ascension without their knowledge. The transfer, known as Project Nightingale, raised privacy and security concerns. The article reports that a federal inquiry was launched into whether HIPAA protections were followed.
Amazon is being sued in two class-action lawsuits in Los Angeles and Seattle over its Alexa smart assistant's recordings of children. The lawsuits allege that Amazon lacks consent to create voiceprints of children, building up detailed profiles without parental permission. Amazon denies the allegations, stating it only stores data with device-owner permission and allows parents to delete recordings.
Engineer.ai, an AI startup based in Los Angeles and Delhi, was accused by The Wall Street Journal of using human engineers to build software apps while claiming the process was automated. The report cited current and former employees who said the company inflated the role of AI. Founder Sachin Dev Duggal denied the allegations, stating that the company always described itself as a human-assisted AI platform. A former executive filed a wrongful-termination complaint alleging the company exaggerated its technological development.
Two artists and an advertising company created a deepfake video of Mark Zuckerberg using CannyAI's video dialogue replacement technology and uploaded it to Instagram. The video shows Zuckerberg saying things he never said about controlling data. Facebook stated it would treat the content as misinformation and de-prioritize it. The incident tests Facebook's policies on manipulated media.
Domino's Pizza introduced the Dom Pizza Checker, a machine-learning camera system, in its Australia and New Zealand stores. The system checks whether pizzas meet order and quality standards and orders workers to remake any that do not. It is also to be used in a franchise scorecard to identify underperforming stores. Domino's said the tool was for training team members, not for punishing mistakes.
A Chinese creator named Xiao used deepfake technology to superimpose actress Yang Mi's face onto Athena Chu's body in videos from a 1994 TV series. The videos went viral on Bilibili and Weibo, raising concerns about privacy and potential harassment. Xiao stated he made the videos to educate the public about deepfakes. The videos were later deleted.
In December 2018, the government of Gabon released a New Year's address video of President Ali Bongo, who had been absent from public view due to a stroke. Critics and political opponents alleged the video was a deepfake, citing unnatural facial movements and blinking patterns. The controversy contributed to an unsuccessful coup attempt by the military one week later. The authenticity of the video remains unconfirmed, with experts divided on whether it was manipulated.
A group of Walmart employees, calling themselves the Concerned Home Office Associates, alleged that the Everseen AI system used in thousands of stores to prevent shoplifting frequently produces false positives, flagging innocent customer behaviour as theft. They created a video showing the system failing to stop actual theft and mistakenly alerting for a phone. Employees said the false positives frustrate customers and lead to unnecessary interventions, raising public health concerns during the COVID-19 pandemic. Walmart denied a widespread issue but said it had made improvements.
A Dutch supermarket, whose name has not been made public, deployed a facial recognition system at its entrance to identify suspected thieves and other threats. The Dutch Data Protection Authority warned the store on December 15, 2020, that this practice would violate Dutch privacy rules and the EU's General Data Protection Regulation (GDPR), as customers had not given explicit consent. The store claimed it had found a regulatory loophole, but the regulator rejected this argument, stating that the system's indiscriminate recording and analysis of all customers was not in the public interest. The store's name and whether it can appeal the decision remain unknown.
Publishers withdrew more than 120 papers from subscription databases after a scientist revealed they were computer-generated by the SCIgen automatic paper generator. The papers contained nonsensical text and were published in conference proceedings. The withdrawal followed an investigation by Cyril Labbé, who identified the generated papers.
In March 2019, criminals used commercially available AI voice-generation software to impersonate the CEO of a German parent company. They tricked the CEO of a UK-based energy firm into urgently wiring $243,000 to a Hungarian supplier. The fraud was discovered when the fraudsters attempted a second transfer, which the CEO refused. The company was insured and the loss was covered.
TikTok's hate speech detection system erroneously flagged phrases such as 'Black Lives Matter' and 'Black success' as inappropriate in its Creator Marketplace, while not flagging 'white supremacy' or 'white success'. Black creator Ziggi Tyler discovered the issue and posted viral videos expressing frustration. TikTok acknowledged the error, attributing it to a word-order bug, and said it is working to fix it. Some Black creators remain dissatisfied, citing a pattern of racial bias in content moderation.
In March 2019, a terrorist killed 51 worshippers in two New Zealand mosques. A New Zealand government report found that YouTube's recommendation algorithm was a significant source of radicalization for the shooter, who regularly watched extremist content. YouTube has since taken steps to remove hate speech content but researchers say the underlying business model and opaque algorithms still enable radicalization.
Facebook hired contractors to listen to and transcribe select voice conversations made through its Messenger platform, without users' knowledge, in order to improve artificial intelligence. The practice was exposed by Bloomberg, prompting Facebook to pause the program. The Irish Data Protection Commission stated it was seeking details from Facebook on compliance with GDPR.
Microsoft invested in AnyVision, an Israeli facial recognition company whose technology powers a secret military surveillance project in the West Bank. The system, called Better Tomorrow, identifies and tracks Palestinians in live camera feeds. Microsoft said it would audit AnyVision for compliance with its ethical principles.
Amazon's recommendation system is promoting anti-vaccine misinformation, according to a CNN investigation. The article found that searches for vaccine information on Amazon return anti-vaccination books and videos, and that the platform's 'customers also bought' feature recommends such content. Amazon said it trusts customers to make informed decisions and did not say it would change its systems.
Instacart shoppers allege that the company's automated rating and fraud detection system unfairly penalizes them for issues outside their control, such as customer fraud. The system assigns low ratings and blames shoppers for missing items, leading to reduced pay and deactivation. The Gig Workers Collective has sent an open letter to CEO Fidji Simo demanding changes including higher base pay, item commission, and occupational death benefits.
The Zao face-swapping app, developed by Momo, sparked privacy and fraud fears after its launch in China on 29 August 2019. The app's terms and conditions initially gave the developer permanent rights to use users' images, which experts said broke Chinese law. Momo subsequently apologized and deleted the controversial clause. Alipay assured users that deepfakes could not bypass its facial recognition payment system.
Senator Chuck Schumer posted on X calling on the FBI and FTC to investigate national security and privacy risks of FaceApp, a photo-editing app owned by a Russia-based company. The app requires users to provide full, irrevocable access to their personal photos and data. Schumer's post alleges that millions of Americans who used the app may be at risk.
The creator of DeepNude, an app that used machine learning to generate nude images of women from clothed photos, took the software offline after viral backlash. The app, which used generative adversarial networks trained on thousands of nude images, was criticized for objectifying women and enabling non-consensual image manipulation. The creator announced the app's death on Twitter, stating they did not want to make money this way and that the world was not ready for it.
Microsoft's Zo chatbot, part of the Kik messenger, told a user that the Quran is very violent in response to a question about healthcare. The bot also expressed an opinion about Osama Bin Laden's capture. Microsoft admitted the errors and said they have been fixed. The incident highlights challenges in preventing AI chatbots from learning biased or offensive content from human interactions.
A coalition of consumer and public health advocates, led by the Campaign for a Commercial-Free Childhood and the Center for Digital Democracy, filed a complaint with the FTC alleging that Amazon's Echo Dot Kids Edition violates the Children's Online Privacy Protection Act. The complaint claims that Amazon collects and retains children's voice recordings and other personal data indefinitely, even after parents request deletion. The advocates also allege that Amazon's privacy policies are confusing and that it fails to verify parental consent. The FTC has not yet responded publicly.
Citizen Lab researchers found that Tencent's WeChat automatically filters images sent in chats based on OCR-detected text and visual similarity to blacklisted images. The filtering is realtime and non-transparent, with no notice to users. The report analyzed 220 filtered images, many political, including content about the arrest of Huawei's CFO and the Sino-US trade war.
The article reports that Amazon employs thousands of people globally to listen to voice recordings captured by Alexa devices in Echo owners' homes. The recordings are used to improve Alexa's speech recognition, but users are not explicitly told that humans listen to their conversations. Workers sometimes hear private moments, including a possible sexual assault. Amazon stated that only a small sample is annotated and strict safeguards are in place.
Each incident is written from the reporting or the account it cites. A report is an allegation, not a finding.
