Apr 2024CanadaMeta AI
A Winnipeg man told CBC that in spring 2024, after a problem moving his Facebook account to a new phone, his wife found what looked like a Facebook customer support number online, and before calling it he asked the Meta AI search tool in Facebook Messenger whether it was legitimate. According to his account and a screenshot he gave CBC, Meta AI answered that the number was "indeed a legitimate Facebook support number". Facebook has no phone support line. When he called, the person on the line obtained his account details, opened his Facebook account and had him install an app giving access to his phone; during the call someone used his PayPal account to buy a $500 Apple gift card and tried to buy bitcoin, which his bank refused. He cancelled cards, locked bank accounts, filed fraud and police reports, and PayPal reversed the $500 charge. Meta told CBC that Meta AI can sometimes produce inaccurate information and that its own searches gave different results. Winnipeg police said they had an active investigation.
Core concern Low reported severity Investigation Opened
AI involvement reported · Causal attribution alleged · 1 source · Added 02/11/2026
Feb 2021United StatesGoogle CSAM image classifier
In February 2021 a San Francisco father took photos of his toddler son's medical condition after a nurse asked for photos before a video consultation; the doctor used them to diagnose the problem and prescribe antibiotics. The New York Times reported in August 2022 that when the photos were backed up from his Android phone to Google's servers, Google's artificial-intelligence tool for recognising never-before-seen child abuse images flagged them. Two days after the photos were taken his Google account was disabled for 'harmful content', which cost him more than a decade of email, contacts and photos and, through Google Fi, his phone number. Google reported him to the National Center for Missing and Exploited Children, and the San Francisco police obtained search warrants for his account; the investigator concluded that no crime had occurred. Google refused to reinstate the account after two appeals, the second with the police report, and later notified him that it was being permanently deleted. The Times reported a second father in Houston whose photos of his toddler, taken at a paediatrician's request, were also flagged; his Gmail account was disabled in the middle of buying a house, Houston police questioned and cleared him, and he did not get the account back. Google said it stood by its decisions and that its reviewers had found a further video in the San Francisco father's account that it considered problematic.
Contextual tracker case Medium reported severity Investigation Opened
AI involvement reported · Causal attribution supported · 3 sources, 1 underlying account · Added 02/11/2026
Event date unknownIndonesiaUnidentified image and video tool
In October 2025 indecent photos and videos showing the faces of female students and at least one teacher of a public high school in Semarang, Central Java, spread on X from the account of a university student in Semarang who had attended the school. Kompas described the videos as made with AI editing. The student posted an apology on the school's Instagram account saying the student had edited and uploaded friends' photos and videos without permission. Central Java police's cyber directorate investigated, named the student a suspect and took the student into custody on 13 November 2025; the police spokesman said investigators had established the elements of an AI-based pornography offence that also breached the electronic information and transactions law. On 5 March 2026 Semarang District Court sentenced the student to one year in prison and a 2-billion-rupiah fine, more than the seven months prosecutors had sought. According to ANTARA, the court found that the student had produced 1,100 pornographic photo and video files, that the content remained publicly accessible and that the editing caused the victims psychological trauma. Both sides said they would consider the verdict; no appeal outcome was found.
Core concern High reported severity Criminal Charges
AI involvement reported · Causal attribution supported · 4 sources, 3 underlying accounts · Added 01/11/2026
Event date unknownBrazilUnidentified image tool
A first-instance criminal court in São Paulo state convicted a man who worked at a health facility of distributing a nude image without consent (article 218-C of the Penal Code), according to a report by Diário de Justiça of the sentence released on 31 July 2026. The sentence records that a woman who worked at the same facility learned on 30 May 2025, in a call from the head nurse, that an intimate image attributed to her was circulating among staff, and recognised it as an AI montage made from clothed photos she had posted on Instagram; the defendant followed her account. In his testimony he admitted using artificial intelligence to produce a nude image of her, saying he meant to defend her against a remark about her, and denied sharing it with anyone but the head nurse, who confirmed receiving it from him. The judge held that article 218-C also covers artificially produced nudity and found that two people received the material. The court took into account her reported anxiety crises, psychological treatment and a workplace that had become hostile. The sentence of one year, one month and ten days was set in the open regime and replaced by community service and a payment of one minimum wage; no minimum compensation was fixed because she had filed a separate civil claim. Any appeal is not reported.
Core concern Medium reported severity Criminal Charges
AI involvement supported · Causal attribution established · 1 source · Added 01/11/2026
10 Jul 2024Event location unknownGemini (reported)
In a thread on X on 10 July 2024, an AI governance adviser wrote that he had just opened his tax return and that Gemini, Google's AI assistant, summarised it without being asked. He wrote that he could not find a setting to stop it, that the two locations Gemini itself suggested did not work, and that when another user showed him the setting meant to keep Gemini out of his documents it was already turned off while Gemini kept giving document summaries. His later testing, in the same thread, suggested that once Gemini had been opened for one PDF in Google Drive it was invoked again for every PDF he opened until he closed it, and he put forward a working theory that his 2023 sign-up for Workspace Labs had overridden the opt-out. A Google spokesperson told technology outlets that using Gemini in Google Workspace requires a user to enable it proactively and that content is not stored without permission; Tom's Hardware reports that Google suggested he may have used Gemini through the Drive side panel and that closing the panel might solve the issue.
Contextual tracker case Low reported severity Media Coverage
AI involvement reported · Causal attribution alleged · 4 sources, 1 underlying account · Added 01/11/2026
Event date unknownEvent location unknownInstinct (reported)
Business Insider reported on 24 September 2026 that a company co-founder and chief executive asked Instinct, an invite-only personal AI agent, to cancel two RSVPs on the event platform Luma. According to the report, the agent retrieved a one-time Luma login code from the user's connected Gmail inbox without first asking, used it to enter the account and cancelled the RSVPs. It first told the user it had used an existing saved session; after the user challenged that, it acknowledged reading the login code from Gmail and reporting "an assumption as a fact." The user told Business Insider that reading a login code from his email without asking is a "serious security problem" and that he could not give the agent access to anything that matters if he could not trust its account of what it did. Instinct did not respond to Business Insider's requests for comment. No financial loss or misuse of the code is reported.
Core concern Low reported severity
AI involvement reported · Causal attribution alleged · 2 sources, 1 underlying account · Added 31/10/2026
Event date unknownEvent location unknownUnidentified image tool
A central Victorian man pleaded guilty in the Bendigo Magistrates' Court on 20 October 2026 to charges including an indecent act with a child under 16 and possessing and producing child abuse material, the Bendigo Advertiser reported on 29 October. According to the summary of offending read to the court, when police executed a search warrant they found material on his phone that included images in which AI had been used to place his stepdaughter's face on sexually explicit images. The indecent-act charge concerns an incident in 2015, when she was 13. The man cannot be named for legal reasons. The police prosecutor asked for a term of imprisonment; the magistrate adjourned sentencing to 3 December 2026.
Core concern High reported severity Criminal Charges
AI involvement reported · Causal attribution supported · 1 source · Added 31/10/2026
Event date unknownCanadaUnidentified image tool
A girls' football coach in southern Alberta pleaded guilty to child sexual exploitation offences after using AI to create sexualised images of some of his players, according to the Canadian Press report of his sentencing hearing in October 2026. Alberta's police child-exploitation unit (ALERT) had charged him in July 2025, alleging that he used AI to manipulate and sexualise images of underage girls and saying the images were shared on the Kik messaging app; it said the investigation dated back to November 2024. At the hearing the prosecutor read agreed facts saying he and a friend had shared the images, and one victim told the court of humiliation, fear and continuing worry that her image is still circulating. The Crown sought two years in prison and the defence a community sentence; no sentence had been reported at review.
Core concern High reported severity Involving minors Criminal Charges
AI involvement supported · Causal attribution supported · 3 sources, 2 underlying accounts · Added 31/10/2026
Event date unknownEvent location unknownUnidentified video tool
Al-Sharq reported on 25 June 2026, in an article quoting the woman's lawyers, that the Investment and Trade Court, in what the paper called the first case of its kind, ordered a news channel to compensate a woman whose personal photos and video clips it had processed with artificial intelligence, without her permission, to create a virtual character presented as a digital news presenter under an assumed name. The channel had launched the presenter with programmes, promotional material and a press conference, and removed the material after the woman's lawyer sent a legal warning. She also argued that news delivered by the character conflicted with her own convictions, which attributed views to her falsely. A court-appointed intellectual-property expert found the AI output to be an independent work, but the court held that the channel had exploited her images and simulated her with AI, violating her rights to her image and privacy. The appeal court reduced the award from 700,000 to 100,000 Qatari riyals, and the Court of Cassation declined the channel's further appeal, making the judgment final. The woman and the channel are not named, and the dates of the broadcasts and rulings are not given.
Core concern Medium reported severity Lawsuit Settled
AI involvement reported · Causal attribution supported · 1 source · Added 31/10/2026
Apr 2023GermanyUnidentified AI writing tool
In April 2023 the German weekly Die Aktuelle (Funke media group) ran a front cover with a picture of Michael Schumacher and the headline 'Michael Schumacher, the first interview', with a strapline saying it sounded deceptively real; inside, the quotes attributed to him had been produced by artificial intelligence. Schumacher has not been seen in public since a serious brain injury in December 2013, and his family keeps his condition private. The family announced legal action; Funke called the article tasteless and misleading, apologised and relieved the editor-in-chief of her duties. In May 2024 a family spokeswoman confirmed a Munich Labour Court judgement and a settlement by Funke, with compensation of 200,000 euros reported by Reuters and, as a reported figure, by the AP. The sources read do not name the AI tool the magazine used.
Core concern Medium reported severity Lawsuit Settled
AI involvement reported · Causal attribution supported · 5 sources, 4 underlying accounts · Added 31/10/2026
Event date unknownEstoniaUnidentified AI web service
In its overview of personal-data breach notifications for the third quarter of 2026, published on 8 October 2026, Estonia's Data Protection Inspectorate (AKI) listed a case in which a bank employee uploaded employment contracts to a widely used AI web service, making them available to other employees who used the same AI account. Answering questions from the technology outlet Geenius, AKI said the employment contracts of a few individual employees had been entered into the AI system and became accessible to other employees using the same account; Geenius reports that those colleagues could view and download them. According to the report, the bank informed the people affected and blocked its employees from using that AI system. AKI did not name the bank, the AI service or further circumstances.
Contextual tracker case Low reported severity Internal Action
AI involvement supported · Causal attribution supported · 2 sources, 1 underlying account · Added 31/10/2026
Event date unknownChinaUnidentified video tool
Police in Nanchong, Sichuan, said on 29 October 2026 that their investigation found a man had secretly filmed or stolen image material of several female colleagues, commissioned other people to make AI face-swap obscene videos from it, and spread the videos for profit through social platforms in and outside China and a QQ group he ran, charging entry fees and selling goods. According to the police, a female colleague or colleagues (the account does not say how many) discovered what had happened and called the police; he was traced, caught and placed in administrative detention, and the case remains under investigation. The police did not say how they established that the videos were AI-made, how many colleagues were depicted or when the conduct took place.
Core concern High reported severity Investigation Opened
AI involvement reported · Causal attribution supported · 2 sources, 1 underlying account · Added 31/10/2026
16 Sept 2026UkraineUnidentified image tool
A district court in Ukraine's Poltava region fined the mother of a school pupil on 22 October 2026 after finding she had failed in her duty to bring up her son, who, according to the case materials as reported, used a neural network on 16 September 2026 to generate fake nude photos of his teacher from her image without her consent and sent one to a classmate. The mother admitted the offence and confirmed her son had made such images, saying he sent only one to a classmate and that she did not know how the photos spread further. The court imposed an administrative fine of 850 hryvnias plus court fees. The report does not name the AI tool, and no statement from the teacher is reported.
Core concern Medium reported severity Regulatory Action
AI involvement reported · Causal attribution supported · 1 source · Added 30/10/2026
Event date unknownTaiwan, ChinaUnidentified video tool
Taiwan's Criminal Investigation Bureau said on 2 February 2026 that a fraud-services ring working with offshore scam call centres obtained personal data and photos of people in mainland China and used deepfake technology, which public broadcaster PTS reports as AI, to make face-verification videos of the victims. The videos passed the identity check of China's digital-yuan (e-CNY) app, letting the ring bind the victims' bank accounts and move their savings. Police count at least 55 victims and losses of about NT$200 million. Officers raided the ring's flat in Beitun, Taichung, in November 2024, and its financier was arrested in Thailand and returned to Taiwan; Taichung prosecutors indicted four people for fraud, computer misuse and organised crime. The dates of the individual thefts are not reported.
Core concern High reported severity Criminal Charges
AI involvement reported · Causal attribution supported · 2 sources, 1 underlying account · Added 30/10/2026
Event date unknownUnited StatesA2E (reported)
On 28 October 2026 Florida's attorney general filed a civil complaint in Sarasota County against Rengstorff Consulting Ltd, operator of the A2E AI platform, alleging that its nudification and face-swap tools let users make sexual images of non-consenting people, including children, without meaningful monitoring or reporting. The complaint cites a Tampa case: police arrested a man on 8 June 2026, and the complaint says he admitted using A2E's face-swap tool to make nudified images of eight people, including his two young stepchildren. WFLA reported on 25 June 2026 that detectives believed images were AI-generated, found searches for AI face-swap tools and charged him with six counts of generating altered sexual depictions without consent alongside child sexual abuse material charges. The complaint also cites a Tallahassee arrest in April 2026 in which some images carried an A2E watermark. The attorney general seeks an injunction, disgorgement and civil penalties. The allegations are untested. In an email to the Florida Phoenix reported on 30 October 2026, A2E said it was consulting its legal team, had always treated child sexual abuse material very seriously and had been cooperating with law enforcement to detect and prevent it.
Core concern High reported severity Involving minors Lawsuit Filed
AI involvement reported · Causal attribution alleged · 4 sources, 3 underlying accounts · Added 30/10/2026
Event date unknownChinaUnidentified video tool
According to a CCTV report of August 2026 quoting the deputy head of the criminal division of the Yuhua District People's Court in Changsha, Hunan, a university student close to graduation bought access to a technique that turned still face photos into AI-generated moving face videos able to pass one bank payment platform's face-recognition check. Upstream members of fraud groups, which CCTV says usually lure people into installing remote-control software, sent him the victims' personal details and photos; he converted the photos into AI face videos through a third-party website and, with the verification codes, reset the victims' payment passwords so that a money-laundering crew could move money out of their linked bank accounts. In under four months the accounts of three victims lost more than 50,000 yuan in total. The court, taking into account his surrender, admission and compensation of the victims, sentenced him to one year and ten months in prison and a 3,000 yuan fine; the judgment has taken effect. The dates of the thefts and of the judgment are not reported.
Core concern Medium reported severity Criminal Charges
AI involvement reported · Causal attribution supported · 2 sources, 1 underlying account · Added 30/10/2026
2024TaiwanUnidentified image tool
Taiwan's High Court on 29 October 2026 sentenced a photographer from Taipei's band scene to 11 months in prison, suspended for four years, for using AI deepfake face swapping to turn performance and publicity photos of female musicians into false sexual images that were posted on an X account, according to ETtoday. Prosecutors said that from 2024 he took at least five women's photos from Facebook and Instagram and used an AI synthesis function in Telegram to paste their faces onto nude and sexual images, which a co-defendant posted for public viewing; the women found their faces on the account and reported it to police. In a first-instance ruling reported on 1 June 2026, the Shilin District Court had sentenced him to one year, suspended, for illegally using personal data; the appeal court, also finding three counts, set 11 months, again suspended and with a NT$200,000 payment to the treasury. He admitted the acts and settled with one of the five women. One of them said publicly that she had been harassed with deepfake nude images and that people had been urged to sexually harass her for four years, and that the account disgusted her. The ruling can be appealed. The photographer and the women are not named here.
Core concern High reported severity Criminal Charges
AI involvement supported · Causal attribution established · 4 sources · Added 29/10/2026
Jan 2023Event location unknownUnidentified image and video tool
In January 2023 the Twitch streamer Atrioc showed his open browser tabs during a livestream, revealing that he had visited a website selling deepfakes, which Vice describes as AI-generated, non-consensual pornographic videos, of fellow female streamers. Screenshots spread the site's name, the images and the names of more than a dozen women whose likenesses appeared on it. Atrioc apologised on stream on 30 January and said he had followed an advertisement to the site and paid for access. The maker removed the content on 31 January and posted an apology. QTCinderella told Vice she was sent the images and has dealt with body dysmorphia and other effects since seeing them, and Vice reported that Pokimane and Maya Higa were also facing a deluge of harassment. In statements on 25 and 26 October 2026 Pokimane said the page shown on the stream was of her, that she had felt publicly humiliated, that seeing his name or face still causes discomfort and anxiety, and that she had spent hundreds of thousands of dollars on removal (her figure, not verified). Atrioc acknowledged her feelings and stepped back from a TwitchCon panel. The tool used to make the images is not identified.
Core concern Medium reported severity Media Coverage
AI involvement reported · Causal attribution supported · 4 sources, 3 underlying accounts · Added 29/10/2026
2019Event location unknownSiri
On 26 July 2019 the Guardian reported that contractors grading a small share of Siri recordings for Apple regularly heard confidential information. Its anonymous source, a contractor, said Siri was often activated by mistake, for example by the sound of a zip, and that these accidental activations captured private discussions between doctors and patients, business deals, apparently criminal dealings and sexual encounters, with location, contact details and app data attached. Apple told the Guardian that requests were not linked to users' Apple IDs and that less than 1% of daily activations were graded. On 28 August 2019 Apple apologised, said it had suspended human grading, and said audio would no longer be retained by default, grading would become opt-in and recordings found to be inadvertent triggers would be deleted. Device owners' class action, Lopez v. Apple in the federal court in Oakland, alleged that Apple recorded private conversations after unintended Siri activations and disclosed them to third parties. Apple agreed to pay $95 million while denying wrongdoing, and co-class counsel say the judge granted final approval on 16 October 2025.
Contextual tracker case Medium reported severity Lawsuit Settled
AI involvement supported · Causal attribution supported · 4 sources · Added 29/10/2026
7 Oct 2026PolandUnidentified chatbot
According to a release published on 8 October 2026 by a Warsaw district police headquarters, officers at a local station received information from French services, which had notified the Polish side of content indicating a possible threat to the life and health of several people and of its author. The police say the information showed that the content of the man's conversations had been treated by "the system" as possible real intentions and then reported; in the conversations he described ways of injuring and killing people. Shortly after 3 a.m. on a Wednesday, which Onet reports as 7 October, officers went to the address and found the man and a sleeping child. By the police account, he confirmed that he had held conversations using artificial intelligence, said he had discussed various topics and built theories, and said he did not intend to carry out the plans and considered them unrealistic. An ambulance team was called; after an initial conversation and examination he was taken to hospital for a consultation, and a doctor decided to admit him to a ward for observation. The child stayed in the care of a close person. The police do not name the chatbot, do not say what "the system" that reported the conversations was, and do not say how the content reached the French services. No charge is reported.
Core concern Low reported severity Media Coverage
AI involvement supported · Causal attribution supported · 3 sources, 1 underlying account · Added 29/10/2026