{"meta":{"exportedAt":"2026-10-23T08:43:21.920Z","formatVersion":2,"selection":{"q":"bc","system":"","harm":"","context":"","country":"","role":"","relation":"core","evidence":"","year":"unknown","response":"","severity":"","verification":"","view":"incidents","sort":"added"},"totalIncidents":19,"coverage":{"cases":19,"countries":5,"languages":3,"unknownLocation":6,"locationPending":0,"unknownLanguage":0,"unknownDate":19,"lawsuits":0,"regulatory":1,"minors":7,"coreRelations":19,"contextualRelations":0,"mixedRelations":0,"unknownRelations":0,"relationPending":0,"relationUnknown":0},"countingNote":"Distinct public cases in this selection. People counts apply within individual cases only; cross-case person overlap has not been resolved. No population incidence estimate.","affectedCountNote":"Interpret person counts with affectedCountStatus and the reported effects. Unquantified zeros are placeholders, not a measured zero.","source":"AI incidents","publisher":"NOPE","url":"https://nope.net/incidents","license":"CC BY 4.0"},"incidents":[{"id":"2024-beverly-hills-california-school-district-expels-five-middle-school-pupils-over-ai-made-nude-images-of-16-classmates","caseFacts":{"claims":[{"id":"c1","status":"reported","evidence":[{"locator":"'were made aware of the “AI-generated nude photos” of students last week, the district superintendent said in a letter to parents. The superintendent told NBC News the photos included students’ faces superimposed onto nude bodies. The district did not share how it determined the photos were produced with artificial intelligence.'","relation":"supports","source_id":"s1"}],"assertion":"The superintendent told parents that school officials had been made aware of AI-generated nude photos of students the previous week and told NBC News the photos included students' faces superimposed onto nude bodies; the district did not say how it determined the photos were produced with AI.","causal_attribution":"The district attributes the images to AI without disclosing its basis."},{"id":"c2","status":"reported","evidence":[{"locator":"'who authorities say used artificial intelligence tools to create the images and share them with other students.'","relation":"supports","source_id":"s2"}],"assertion":"A Beverly Hills deputy city manager said police were investigating pupils who authorities said used artificial intelligence tools to create the images and share them with other students.","causal_attribution":"City official's statement attributing the images to AI use by pupils."},{"id":"c3","status":"reported","evidence":[{"locator":"'The Beverly Hills Unified School District voted this week to confirm the expulsion of five middle school students' and 'The five students and their victims were in the eighth grade, according to the school district. Sixteen students were targeted,' and 'No arrests have been made and no charges have been brought.'","relation":"supports","source_id":"s3"}],"assertion":"The school board confirmed the expulsion of five eighth-grade pupils in March 2024; the superintendent said 16 students were targeted, and as of that report no arrests had been made and no charges brought.","causal_attribution":"Disciplinary outcome and count as stated by the district."},{"id":"c4","status":"reported","evidence":[{"locator":"'Students and parents told NBC News they were afraid to go to school or send their children to school after the incident,'","relation":"supports","source_id":"s2"}],"assertion":"Students and parents told NBC News they were afraid to go to school or send their children to school after the incident.","causal_attribution":"Reported distress following the circulation of the images."}],"effects":[{"label":"16 pupils depicted in AI-made nude images shared among students, according to the district","claim_id":"c3","direction":"negative"},{"label":"students afraid to go to school","claim_id":"c4","direction":"negative"}],"sources":[{"id":"s1","url":"https://www.nbcnews.com/tech/misinformation/beverly-vista-hills-middle-school-ai-images-deepfakes-rcna140775","kind":"news_report","access":"read","language":"en","translation_note":"Read in English on 2026-10-22 (HTTP 200). Published 27 February 2024; reports the superintendent letter and a police lieutenant.","independence_group":"bhusd-superintendent-letter-and-statements-feb-2024"},{"id":"s2","url":"https://www.nbcnews.com/tech/tech-news/beverly-hills-ai-nude-photos-middle-school-vista-rcna140965","kind":"news_report","access":"read","language":"en","translation_note":"Read in English on 2026-10-22 (HTTP 200). Published 28 February 2024; reports a deputy city manager's letter to parents.","independence_group":"beverly-hills-city-letter-feb-2024"},{"id":"s3","url":"https://www.nbcnews.com/tech/misinformation/beverly-hills-school-expels-students-deepfake-nude-images-rcna142480","kind":"news_report","access":"read","language":"en","translation_note":"Read in English on 2026-10-22 (HTTP 200). Published 8 March 2024; reports the board decision via the Los Angeles Times and minutes, and the superintendent email.","independence_group":"bhusd-expulsion-decision-mar-2024"}],"version":1,"ai_roles":["others_use"],"contexts":["education"],"unknowns":["Which AI tools were used and how the district determined the images were AI-made.","When the images were made and first shared; the district learned of them in February 2024.","Whether the police investigation led to any charge after March 2024."],"geography":{"basis":"Event: a middle school in the Beverly Hills Unified School District, Beverly Hills, California, with Beverly Hills police investigating (NBC News). No court proceeding is reported.","court_countries":[],"event_countries":["US"],"affected_person_countries":["US"]},"publication":{"basis":"Published as a core case (depicted_or_impersonated): the school district that investigated and expelled five pupils describes AI-generated nude images of 16 classmates, and a city official says authorities consider that pupils used AI tools to make and share them. The district's technical basis is not disclosed. No pupil is named.","reviewed_on":"2026-10-22"},"ai_involvement":{"basis":"The Beverly Hills Unified School District, which investigated and expelled five pupils, described the images as AI-generated nude photos that put students' faces on nude bodies, and a city official said authorities consider that pupils used artificial intelligence tools to create and share them (NBC News, February and March 2024). The images depict the 16 targeted pupils (depicted_or_impersonated). The district did not say how it determined that AI was used, and no tool is named.","status":"reported"},"person_relations":["depicted_or_impersonated"]},"name":"Beverly Hills, California: school district says middle school pupils made AI nude images of 16 classmates and expels five","summary":"In February 2024 the Beverly Hills Unified School District told parents that officials at one of its middle schools had been made aware of AI-generated nude photos of students, which the superintendent said put students' faces on nude bodies, and a city official said police were investigating pupils whom authorities said used AI tools to create and share the images. In March 2024 the district confirmed the expulsion of five eighth-grade pupils; the superintendent said 16 students were targeted. Students and parents told NBC News they were afraid to go to school. The district did not say how it determined that AI was used.","incidentKind":"single_event","incidentDatePrecision":"unknown","exposurePattern":"unknown","reportedDate":"2024-02-27","aiSystem":"Generative AI tools used to put classmates' faces on nude bodies, according to the school district and a city official (tools not identified)","aiProduct":"Unidentified image tool","severity":"high","verificationStatus":"credible","harmCategories":[],"harmOutcomes":["exploitation_or_abuse","psychological_distress"],"harmOutcomeSummary":"According to the school district, pupils used AI to put 16 classmates' faces on nude images that were shared among students; students and parents told NBC News they were afraid to go to school.","frameworkFacets":[],"pathwayDirection":"outward","causationStatus":"supported","participantUsersAffectedMin":0,"otherPeopleHarmedMin":16,"affectedCountStatus":"exact","affectedCountEvidence":"Sixteen students were targeted, according to the superintendent's email to the district community (NBC News, 8 March 2024).","victimAgeRange":"minor","jurisdiction":"US","platformType":"other","outcomeType":"internal_action","outcomeStatus":"unknown","primarySourceUrl":"https://www.nbcnews.com/tech/misinformation/beverly-hills-school-expels-students-deepfake-nude-images-rcna142480","primarySourceLabel":"NBC News, 8 March 2024: Beverly Hills middle school expels 5 students after deepfake nude photos incident","firstPublishedAt":"2026-10-22T03:16:26.033755+00:00","updatedAt":"2026-10-22T03:16:26.033755+00:00","scopeVersion":"facts-v3","tags":["ai-generated-sexual-images","sexual-images-of-minors","school","expulsion","depicted-or-impersonated"]},{"id":"2026-toronto-car-dealership-revoked-buy-back-offer-its-undisclosed-ai-chatbot-texted-customer-then-reinstated-it","caseFacts":{"claims":[{"id":"c1","status":"reported","evidence":[{"locator":"'Later in the same text conversation, Quinn made a firm buy-back offer: $27,162.79.' and 'texting that “the team” would consider it, and that “$28,500 sounds reasonable.”' and 'Quinn never revealed its true identity, according to text exchanges examined by CBC News.'","relation":"supports","source_id":"s1"}],"assertion":"A dealership texting agent called Quinn made the customer a firm buy-back offer of $27,162.79, said his $28,500 counter sounded reasonable and set up a meeting to complete the deal; CBC says Quinn never revealed it was AI.","causal_attribution":"Customer account supported by the text exchange CBC examined."},{"id":"c2","status":"reported","evidence":[{"locator":"'a BMW Toronto sales consultant called to revoke the offer, explaining that Quinn wasn’t a real person, but rather an artificial intelligence chatbot that had made the offer in error.' and 'the actual buy-back offer would be, at best, $20,000'","relation":"supports","source_id":"s1"}],"assertion":"A sales consultant revoked the offer, saying Quinn was an AI chatbot that had made it in error, and said the real offer would be at best $20,000.","causal_attribution":"Customer's account; the dealership's sales manager confirms the bot's error."},{"id":"c3","status":"reported","evidence":[{"locator":"'I was shocked, I was astounded, like my jaw was on the floor.' and '“I feel embarrassed, and I feel angry that I’ve been negotiating with this bot,”'","relation":"supports","source_id":"s1"}],"assertion":"The customer says he was shocked and astounded, and embarrassed and angry that he had been negotiating with a bot.","causal_attribution":"The customer's own account."},{"id":"c4","status":"reported","evidence":[{"locator":"'“The AI bot ran with that, misunderstood the message,” he said.' and 'Quinn was never programmed to independently negotiate contracts, only to relay human-generated buy-back offers.' and 'After CBC News reached out to BMW Toronto, it reinstated Quinn’s original offer of $27,162.79.'","relation":"supports","source_id":"s1"}],"assertion":"The sales manager said a human employee's miscommunication led the bot to take the amount owed as the buy-back price, that the bot was not programmed to negotiate, and that the dealership reinstated the offer after CBC's inquiry.","causal_attribution":"Dealership's explanation and CBC's report of the reinstatement."}],"effects":[{"label":"chatbot offer revoked and cut by more than $7,000 until reinstated; distress at undisclosed bot","claim_id":"c3","direction":"negative"},{"label":"original offer reinstated and accepted after CBC inquiry","claim_id":"c4","direction":"positive"}],"sources":[{"id":"s1","url":"https://www.cbc.ca/news/business/ai-chatbot-bmw-dealership-9.7230226","kind":"news_report","access":"read","language":"en","translation_note":"Read in English by the research agent (an AI) on 2026-10-22; no translation needed. CBC News report by Sophia Harris, posted 11 Jun 2026; HTTP 200.","independence_group":"cbc-bmw-toronto-june-2026"}],"version":1,"ai_roles":["institutional_use"],"contexts":["finance","everyday_life"],"unknowns":["The exact date of the text exchange (CBC says he decided the month before its June 2026 report to sell the car).","Which vendor supplied the chatbot and whether it is generative or scripted."],"geography":{"basis":"The dealership is BMW Toronto and CBC describes the customer as a Toronto man.","court_countries":[],"event_countries":["CA"],"affected_person_countries":["CA"]},"publication":{"basis":"Published as a core case (communicated_with): a dealership's undisclosed AI texting chatbot made a customer a firm buy-back offer that staff revoked as the bot's error, leaving him devastated until the dealership reinstated it after CBC's inquiry.","reviewed_on":"2026-10-22"},"ai_involvement":{"basis":"The customer negotiated by text with an AI chatbot the dealership deployed; the chatbot made a firm buy-back offer that staff revoked as the bot's error, which the dealership's sales manager confirmed ('The AI bot ran with that, misunderstood the message'). The revoked offer and the customer's distress follow from the chatbot's own offer; the manager attributes the error partly to a staff miscommunication. The bot's output was not a fixed relay of a human offer: in the texts it expressed sympathy, asked about the car, answered the counter-offer ('$28,500 sounds reasonable') and proposed a meeting time, and the dealership says it misinterpreted a staff member's message. Whether it runs on a generative model is not stated.","status":"reported"},"person_relations":["communicated_with"]},"name":"Toronto: BMW dealership revoked a buy-back offer its undisclosed AI chatbot had texted a customer, then reinstated it after CBC asked","summary":"CBC News reported in June 2026 that a Toronto man selling his car back to the BMW Toronto dealership negotiated by text with 'Quinn', who made a firm buy-back offer of $27,162.79 and proposed a meeting to complete the deal. A sales consultant then phoned to say Quinn was an AI chatbot that had made the offer in error and that the real offer would be at best $20,000. The man says he was shocked and devastated and felt embarrassed and angry that he had been negotiating with a bot without knowing it. The dealership's sales manager said the bot misread the amount the customer owed as the buy-back price after a staff miscommunication; after CBC contacted it, the dealership reinstated the original offer, which the man accepted.","incidentKind":"single_event","incidentDatePrecision":"unknown","exposurePattern":"single_interaction","reportedDate":"2026-06-11","aiSystem":"An AI texting chatbot presenting as 'Quinn' that BMW Toronto used to handle an online buy-back inquiry; the dealership says it was meant only to relay human-generated offers (vendor not named)","aiProduct":"Unidentified chatbot","severity":"low","verificationStatus":"credible","harmCategories":[],"harmOutcomes":["psychological_distress"],"harmOutcomeSummary":"A customer says a dealership's AI chatbot texted him a firm buy-back offer that staff then revoked as the bot's error, cutting it by more than $7,000 until CBC intervened; he says he was devastated and embarrassed to learn he had been negotiating with a bot.","frameworkFacets":[],"pathwayDirection":"inward","causationStatus":"supported","participantUsersAffectedMin":1,"otherPeopleHarmedMin":0,"affectedCountStatus":"exact","affectedCountEvidence":"One customer, interviewed by CBC.","victimAgeRange":"adult","jurisdiction":"CA","platformType":"other","outcomeType":"media_coverage","outcomeStatus":"resolved","primarySourceUrl":"https://www.cbc.ca/news/business/ai-chatbot-bmw-dealership-9.7230226","primarySourceLabel":"CBC News, 11 Jun 2026: Dealership revoked offer to buy back customer's BMW, blaming wayward AI chatbot","firstPublishedAt":"2026-10-22T03:14:15.002018+00:00","updatedAt":"2026-10-22T03:14:15.002018+00:00","scopeVersion":"facts-v3","tags":["customer_service_chatbot","undisclosed_ai","car_dealership","canada"]},{"id":"2026-staffordshire-man-jailed-after-police-say-he-used-ai-to-make-indecent-images-of-women-and-a-girl-posted-on-telegram","caseFacts":{"claims":[{"id":"c1","status":"reported","evidence":[{"locator":"'A man used artificial intelligence (AI) platforms to create indecent images of women and girls, Staffordshire Police said.' ... 'posted a series of AI-generated pornographic images on closed Telegram groups.'","relation":"supports","source_id":"s1"},{"locator":"'Staffordshire Police say the defendant had posted a series of pornographic and indecent images of women and a girl on 'closed' Telegram groups that had been run through AI.'","relation":"supports","source_id":"s2"}],"assertion":"Staffordshire Police said the man used artificial intelligence platforms to create indecent images of women and girls and posted a series of AI-generated pornographic images in closed Telegram groups; The Sentinel reports the police account as images of women and a girl that had been run through AI.","causal_attribution":"Staffordshire Police's account as relayed by BBC News and The Sentinel (one police chain); the platforms are not named and the technical basis of the AI finding is not described."},{"id":"c2","status":"reported","evidence":[{"locator":"'he called for the women depicted in the images to be raped and assaulted, investigators found.'","relation":"supports","source_id":"s1"},{"locator":"'he called for the women depicted in the images to be raped and'","relation":"supports","source_id":"s3"}],"assertion":"Investigators found that in some captions he called for the women depicted in the images to be raped and assaulted.","causal_attribution":"Investigators' finding as relayed by BBC News and the Daily Express (one police chain)."},{"id":"c3","status":"reported","evidence":[{"locator":"'We turned up at his home and arrested him after the images were reported to us.' ... 'after admitting the offences on 27 August.'","relation":"supports","source_id":"s3"},{"locator":"'He was jailed for two years and two months after pleading guilty at Stafford Crown Court on 27 August.' ... 'He was charged with eight counts of publishing an obscene article and one count of permitting the taking of a pseudo-photograph of a child. He admitted all of the charges.'","relation":"supports","source_id":"s1"}],"assertion":"After the images were reported, police arrested him at his home and seized his devices; he admitted eight counts of publishing an obscene article and one count of permitting the taking of a pseudo-photograph of a child, and was jailed for two years and two months after pleading guilty at Stafford Crown Court on 27 August 2026.","causal_attribution":"Police statement and court outcome as reported by BBC News and the Daily Express."},{"id":"c4","status":"reported","evidence":[{"locator":"'This type of crime can have a profound impact on victims'","relation":"supports","source_id":"s1"}],"assertion":"The investigating officer said this type of crime can have a profound impact on victims.","causal_attribution":"The investigating officer's general statement about this type of crime, quoted by BBC News."}],"effects":[{"label":"depicted in AI-generated pornographic images posted in closed Telegram groups","claim_id":"c1","direction":"negative"},{"label":"captions called for the women depicted to be raped and assaulted","claim_id":"c2","direction":"negative"}],"sources":[{"id":"s1","url":"https://www.bbc.com/news/articles/cyvzn1zvz10o","kind":"news_report","access":"read","language":"en","translation_note":"Read in English on 2026-10-22 (HTTP 200). BBC News report of 9 September 2026 attributing the account to Staffordshire Police.","independence_group":"staffordshire-police-account"},{"id":"s2","url":"https://www.msn.com/en-gb/crime/general/pervert-used-ai-to-make-indecent-images-of-women/ar-AA2c0JJ4","kind":"news_report","access":"read","language":"en","translation_note":"Read in English on 2026-10-22 through the MSN syndication copy (HTTP 200). The Sentinel report of 11 September 2026 paraphrasing Staffordshire Police.","independence_group":"staffordshire-police-account"},{"id":"s3","url":"https://www.msn.com/en-gb/news/other/pervert-creates-indecent-ai-images-of-women-and-girls-posts-them-on-social-media/ar-AA2bXu8A","kind":"news_report","access":"read","language":"en","translation_note":"Read in English on 2026-10-22 through the MSN syndication copy (HTTP 200). Daily Express report of 10 September 2026 quoting a Staffordshire Police statement.","independence_group":"staffordshire-police-account"}],"version":1,"ai_roles":["others_use"],"contexts":["privacy","justice"],"unknowns":["How many women and girls were depicted, and whether they were identified or told.","Whether the women and girl depicted are real, identifiable people: The Sentinel reports the police account as images that had been run through AI, and police refer to victims, but no source says a depicted person was identified.","How the images of the people depicted were obtained and which AI platforms were used.","When the images were posted.","The date of sentencing (the reports give the plea date and the sentence)."],"geography":{"basis":"Staffordshire Police arrested the man at his home in Staffordshire and the case was heard at Stafford Crown Court (all three sources), so the event and court country is GB. The sources do not say where the depicted people live, so affected_person_countries is empty.","court_countries":["GB"],"event_countries":["GB"],"affected_person_countries":[]},"publication":{"basis":"Published as a core case (depicted_or_impersonated): the investigating police force says the man used AI platforms to create indecent images of women and girls that he posted in closed Telegram groups, with captions calling for the women depicted to be raped, and The Sentinel reports the police account as images of women and a girl run through AI; he pleaded guilty and was jailed. The people depicted are not named or counted, and the convicted man is not named.","reviewed_on":"2026-10-22"},"ai_involvement":{"basis":"Staffordshire Police, which investigated and charged the case, said the man used artificial intelligence platforms to create indecent images of women and girls and posted the AI-generated pornographic images in closed Telegram groups (BBC). The Sentinel reports the police account as images of women and a girl that had been run through AI. The images depict the women and girl (depicted_or_impersonated). The platforms are not named and the technical basis of the police finding is not described beyond the seized devices and the guilty plea.","status":"reported"},"person_relations":["depicted_or_impersonated"]},"name":"Staffordshire: man jailed after police say he used AI platforms to make indecent images of women and a girl and posted them in Telegram groups","summary":"Staffordshire Police said a man used artificial intelligence platforms to create indecent images of women and girls and posted a series of the AI-generated pornographic images in closed Telegram groups, as reported by BBC News, the Daily Express and The Sentinel. Investigators found that in some captions he called for the women depicted to be raped and assaulted. The Sentinel reports the police account as images of women and a girl that had been run through AI. He was arrested after the images were reported to police, pleaded guilty at Stafford Crown Court to eight counts of publishing an obscene article and one of permitting the taking of a pseudo-photograph of a child, and was jailed for two years and two months. No source names or counts the people depicted.","incidentKind":"bounded_series","incidentDatePrecision":"unknown","exposurePattern":"repeated_interactions","reportedDate":"2026-09-09","aiSystem":"Artificial intelligence platforms used to create indecent images, according to Staffordshire Police (platforms not named)","aiProduct":"Unidentified image tool","severity":"high","verificationStatus":"credible","harmCategories":[],"harmOutcomes":["exploitation_or_abuse"],"harmOutcomeSummary":"According to Staffordshire Police as reported by three outlets, AI-generated pornographic images of women and a girl were posted in closed Telegram groups, some captioned with calls for the women depicted to be raped and assaulted.","frameworkFacets":[],"pathwayDirection":"outward","causationStatus":"established","participantUsersAffectedMin":0,"otherPeopleHarmedMin":0,"affectedCountStatus":"unquantified","affectedCountEvidence":"Unquantified: the sources describe the people depicted only as women and girls (BBC) or women and a girl (The Sentinel), with no count, so no minimum is recorded. The convicted man is not counted as harmed.","victimAgeRange":"unknown","jurisdiction":"GB-ENG","platformType":"other","outcomeType":"criminal_charges","outcomeStatus":"resolved","primarySourceUrl":"https://www.bbc.com/news/articles/cyvzn1zvz10o","primarySourceLabel":"BBC News, 9 September 2026: Man who created indecent AI images of girls jailed","firstPublishedAt":"2026-10-22T03:10:18.448466+00:00","updatedAt":"2026-10-22T03:10:18.448466+00:00","scopeVersion":"facts-v3","tags":["ai-generated-images","telegram","obscene-publications","pseudo-photograph","crown-court","staffordshire","guilty-plea","sentencing"]},{"id":"2023-greater-manchester-prosecutors-say-man-used-ai-to-turn-photos-of-real-children-into-abuse-images-sold-to-order-jailed","caseFacts":{"claims":[{"id":"c1","status":"reported","evidence":[{"locator":"used a computer programme and AI technology to alter photographs of real children to create sexual abuse images","relation":"supports","source_id":"s1"},{"locator":"A student who used AI technology \"in the worst possible way\" to turn photographs of real children into \"depraved\" indecent images has been jailed for 18 years.","relation":"supports","source_id":"s2"}],"assertion":"The CPS says the man used a computer programme and AI technology to alter photographs of real children to create sexual abuse images.","causal_attribution":"The prosecuting authority, which examined the evidence from his seized devices, attributes the images to AI technology; the BBC court report repeats it, partly quoting the CPS, so it is one prosecution origin. The tool is not named."},{"id":"c2","status":"reported","evidence":[{"locator":"accepted requests from individuals who wanted him to create explicit images depicting children being harmed both sexually and physically, which he either sold in exchange for money or shared to others for free. Some of these computer-generated images requested used images of real children.","relation":"supports","source_id":"s1"},{"locator":"The court heard he took requests from people who wanted him to create explicit images of children being harmed both sexually and physically.","relation":"supports","source_id":"s2"}],"assertion":"According to the CPS, he accepted requests in encrypted chatrooms to create explicit images of children being harmed, sold or shared them, and some of the requested images used images of real children; he admitted making around 5,000 pounds over 18 months.","causal_attribution":"Prosecution account of how the AI-altered images were commissioned and distributed."},{"id":"c3","status":"reported","evidence":[{"locator":"previously pleading guilty to a total of 16 charges relating to child sexual abuse offences. ... On 28 October 2024, he was sentenced to 18 years' imprisonment with an additional 6 years on extended licence.","relation":"supports","source_id":"s1"},{"locator":"who pleaded guilty to 16 child sexual abuse offences, also encouraged the rape of children in online chatrooms.","relation":"supports","source_id":"s2"}],"assertion":"He pleaded guilty to 16 child sexual abuse offences, including three counts of encouraging the rape of a child under 13, and on 28 October 2024 was sentenced at Bolton Crown Court to 18 years' imprisonment with an additional six years on extended licence.","causal_attribution":"Court outcome; the encouragement offences concern chatroom messages, not the AI images."},{"id":"c4","status":"reported","evidence":[{"locator":"The misuse of emerging technology to create this material is a serious crime which can have a real impact on victims. Real children are being victimised through the creation of this content.","relation":"supports","source_id":"s1"}],"assertion":"The CPS prosecutor said real children are victimised through the creation of such content.","causal_attribution":"Prosecutor's statement of harm; no individual child's experience is described."},{"id":"c5","status":"reported","evidence":[{"locator":"after he was first arrested in June 2023 by Greater Manchester Police, and a number of his digital devices were seized and examined.","relation":"supports","source_id":"s1"},{"locator":"His offences were uncovered when he began speaking to an undercover police officer in May last year.","relation":"supports","source_id":"s2"}],"assertion":"Greater Manchester Police arrested him in June 2023 and examined his devices; the BBC court report says the offences were uncovered when he spoke to an undercover officer.","causal_attribution":"Investigation chronology; the BBC's 'May last year' (2023) is the court report's wording."}],"effects":[{"label":"real children's photographs altered with AI into sexual abuse images sold to order","claim_id":"c1","direction":"negative"}],"sources":[{"id":"s1","url":"https://www.cps.gov.uk/cps/news/man-who-used-ai-technology-create-child-sexual-abuse-images-jailed","kind":"official_statement","access":"read","language":"en","translation_note":"Read in English by the research agent (an AI) on 2026-10-20; HTTP 200. Crown Prosecution Service press release of 28 October 2024; no translation needed.","independence_group":"cps-28-oct-2024"},{"id":"s2","url":"https://www.bbc.com/news/articles/cq6l241z5mjo","kind":"news_report","access":"read","language":"en","translation_note":"Read in English by the research agent (an AI) on 2026-10-20; HTTP 200. BBC court report of the sentencing hearing, quoting the judge, the prosecutor, the police and the CPS.","independence_group":"bbc-court-report-28-oct-2024"}],"version":1,"ai_roles":["others_use"],"contexts":["privacy","justice"],"unknowns":["Which AI tool and computer programme were used.","How many real children were depicted and whether any were identified or informed.","When the 18-month period of selling began; the sources date only the June 2023 arrest and the 2023 and 2024 pleas."],"geography":{"basis":"The case was investigated by Greater Manchester Police and sentenced at Bolton Crown Court, United Kingdom (GB). The images were made and traded online and the BBC says some requests came from abroad; the sources do not say where the depicted children live or where the images were made, so no event or affected-person country is recorded.","court_countries":["GB"],"event_countries":[],"affected_person_countries":[]},"publication":{"basis":"Published as a core case (depicted_or_impersonated): the Crown Prosecution Service says the man used AI technology to alter photographs of real children into sexual abuse images he made to order and sold, and the BBC court report gives the same account of the 2024 sentencing. No child is identified; the defendant is not named in the record.","reviewed_on":"2026-10-20"},"ai_involvement":{"basis":"The Crown Prosecution Service, which brought the case after police examined his seized devices, says the man used a computer programme and AI technology to alter photographs of real children into sexual abuse images that he made to order and sold or shared; the BBC court report of the sentencing gives the same account. The AI-altered depictions of real children are the abuse material at the centre of the offences. The tool is not named in the sources read.","status":"reported"},"person_relations":["depicted_or_impersonated"]},"name":"UK prosecutors say a Greater Manchester man used AI tools to turn photos of real children into sexual abuse images he sold to order; jailed for 18 years in 2024","summary":"The Crown Prosecution Service said on 28 October 2024 that a man from Bolton had been sentenced at Bolton Crown Court to 18 years' imprisonment with an additional six years on extended licence after pleading guilty to 16 child sexual abuse offences. According to the CPS, he used a computer programme and AI technology to alter photographs of real children into sexual abuse images, which he made to order for people in encrypted chatrooms and sold or shared, and some of the requested images used images of real children. The CPS says he admitted making about 5,000 pounds over 18 months. He was arrested by Greater Manchester Police in June 2023. The BBC court report says the offences were uncovered when he spoke to an undercover officer. He also admitted encouraging the rape of children under 13 in chatrooms.","incidentKind":"bounded_series","incidentDatePrecision":"unknown","exposurePattern":"unknown","reportedDate":"2024-10-28","aiSystem":"A computer programme and AI technology that, according to the CPS, the man used to alter photographs of real children into sexual abuse images (tool not named in the sources read)","aiProduct":"Unidentified image tool","severity":"critical","verificationStatus":"credible","harmCategories":[],"harmOutcomes":["exploitation_or_abuse"],"harmOutcomeSummary":"According to the CPS, a man used AI technology to alter photographs of real children into sexual abuse images that he made to order and sold or shared in chatrooms; prosecutors say real children are victimised through such content.","frameworkFacets":[],"pathwayDirection":"outward","causationStatus":"supported","participantUsersAffectedMin":0,"otherPeopleHarmedMin":0,"affectedCountStatus":"unquantified","affectedCountEvidence":"No count: the CPS says some requested images used images of real children but gives no number of children depicted. The '60 characters' in the BBC court report are characters he described, not a count of real children.","victimAgeRange":"minor","jurisdiction":"GB","platformType":"other","outcomeType":"criminal_charges","outcomeStatus":"resolved","primarySourceUrl":"https://www.cps.gov.uk/cps/news/man-who-used-ai-technology-create-child-sexual-abuse-images-jailed","primarySourceLabel":"Crown Prosecution Service, 28 Oct 2024: Man who used AI technology to create child sexual abuse images jailed","firstPublishedAt":"2026-10-20T03:18:07.543351+00:00","updatedAt":"2026-10-20T03:18:07.543351+00:00","scopeVersion":"facts-v3","tags":["ai_altered_images","child_sexual_abuse_material","images_made_to_order","guilty_plea","united_kingdom"]},{"id":"2025-greater-manchester-former-supply-teacher-jailed-over-ai-sexualised-images-of-pupils-and-upskirting","caseFacts":{"claims":[{"id":"c1","status":"reported","evidence":[{"locator":"'used his position as a teacher to find information about pupils so he could locate them on social media and use their own photos to create sexualised images using AI technology'; 'Officers found deepfake images where the image of teenage children had been put onto the body of adults or alongside adults engaging in sexual activity.'","relation":"supports","source_id":"s2"},{"locator":"'used artificial intelligence (AI) to create sexualised fake images from pupils' social media photographs which he shared online'","relation":"supports","source_id":"s1"}],"assertion":"According to the Crown Prosecution Service, the defendant used his position as a teacher to find pupils' details, located them on social media and used AI to create sexualised images from their own photos; officers found deepfake images in which teenage children had been put onto adults' bodies or alongside adults engaged in sexual activity.","causal_attribution":"Prosecution account after a guilty plea; the BBC's narrative matches it but may draw on the same CPS statement, so the claim is treated as one origin."},{"id":"c2","status":"reported","evidence":[{"locator":"'Some of the images were posted on X, where one of his accounts had about 1,000 followers.'","relation":"supports","source_id":"s1"},{"locator":"'created and shared sexualised deepfake images of pupils on social media'","relation":"supports","source_id":"s2"}],"assertion":"The images were shared on social media; the BBC reports some were posted on X, where one of the defendant's accounts had about 1,000 followers.","causal_attribution":"BBC report and the senior crown prosecutor's statement; the X detail and follower count come from the BBC only."},{"id":"c3","status":"reported","evidence":[{"locator":"'On 2 March 2025, police received several reports from concerned parents about two social media accounts where they had seen images of school pupils.'","relation":"supports","source_id":"s2"},{"locator":"'came across an image of her daughter in her school uniform next to graphic sexual images'","relation":"supports","source_id":"s1"}],"assertion":"Police received several reports from concerned parents on 2 March 2025 about two social media accounts showing images of school pupils; the BBC reports a girl's mother found an image of her daughter in school uniform next to graphic sexual images on X.","causal_attribution":"How the offending was discovered; context."},{"id":"c4","status":"reported","evidence":[{"locator":"'The court heard one girl whose social media photograph had been used was left embarrassed and \"weirded out\", while another felt \"disgusted and upset\".'","relation":"supports","source_id":"s1"},{"locator":"'it has had a profound impact on both the victims and wider community'","relation":"supports","source_id":"s2"}],"assertion":"The court heard that one girl whose social media photograph had been used was left embarrassed and 'weirded out', and the senior crown prosecutor said the offending had a profound impact on the victims and the wider community.","causal_attribution":"Victim impact as heard in court and as the prosecutor describes it."},{"id":"c5","status":"corroborated","evidence":[{"locator":"'He pleaded guilty to 21 offences, including taking indecent photographs of children and publishing obscene material.'; '20-year Sexual Harm Prevention Order banning him from working in schools'","relation":"supports","source_id":"s1"},{"locator":"'was jailed at Bolton Crown Court, for six years and was made the subject of a 20-year Sexual Harm Prevention Order to prevent further offending'; 'pleaded guilty at Bolton Crown Court on 28 May 2026'; '4 counts of publishing an obscene article'","relation":"supports","source_id":"s2"}],"assertion":"The defendant pleaded guilty at Bolton Crown Court in May 2026 to 21 offences, including publishing obscene material, and on 15 October 2026 was jailed for six years with a 20-year Sexual Harm Prevention Order barring him from working in schools.","causal_attribution":"The plea and sentence as reported from court by the BBC and stated by the prosecuting authority."},{"id":"c6","status":"reported","evidence":[{"locator":"'taking more than 100 upskirt photographs of schoolgirls, including in his classroom'; 'The court heard 15 girls, aged 13 to 16, were targeted at seven secondary schools in Bolton, Bury and Salford. Police said 12 had been identified.'","relation":"supports","source_id":"s1"}],"assertion":"The defendant also took more than 100 covert upskirt photographs of schoolgirls, including in classrooms; the court heard 15 girls were targeted across the offending and police had identified 12.","causal_attribution":"Non-AI offending in the same prosecution; context for the counts."}],"effects":[{"label":"pupils' social media photos made into sexualised AI images and shared online","claim_id":"c1","direction":"negative"},{"label":"a girl whose photo was used left embarrassed","claim_id":"c4","direction":"negative"}],"sources":[{"id":"s1","url":"https://www.bbc.co.uk/news/articles/ck79p174x3d3o","kind":"news_report","access":"read","language":"en","translation_note":"Read in English on 2026-10-16 (HTTP 200). BBC court report of the 15 October 2026 sentencing, quoting the prosecutor, the judge, the defence and Greater Manchester Police; it also quotes the CPS statement.","independence_group":"bolton-crown-court-sentencing-bbc"},{"id":"s2","url":"https://www.cps.gov.uk/north-west/news/former-supply-teacher-jailed-sexual-offences-including-deepfake-images-and","kind":"official_statement","access":"read","language":"en","translation_note":"Read in English on 2026-10-16 (HTTP 200). Crown Prosecution Service North West press release of 15 October 2026.","independence_group":"cps-north-west-release"}],"version":1,"ai_roles":["others_use"],"contexts":["privacy","education","justice"],"unknowns":["Which AI tool was used and when the AI images were made","How many pupils were depicted in the AI images, as distinct from the upskirt photographs","How long the images stayed online, who saw them and whether the platform removed them"],"geography":{"basis":"Event and affected people: the BBC reports the targeted girls attended seven secondary schools in Bolton, Bury and Salford, Greater Manchester, where the defendant worked. Court: Bolton Crown Court.","court_countries":["GB"],"event_countries":["GB"],"affected_person_countries":["GB"]},"publication":{"basis":"Published on the prosecuting authority's account after a guilty plea: the Crown Prosecution Service says the former supply teacher used AI to create sexualised images from pupils' social media photos and shared them, and officers found the deepfake images on his devices; the BBC court report says the same and that a girl whose photo was used was left embarrassed. The plea and sentence are reported by both. The pupils are minors and are not identified; the defendant is not named because his name adds nothing to the record.","reviewed_on":"2026-10-16"},"ai_involvement":{"basis":"The Crown Prosecution Service, which prosecuted the case to a guilty plea, says the defendant used pupils' own social media photos to create sexualised images using AI technology, and that officers found deepfake images on his seized devices placing teenage children onto adults' bodies or alongside sexual activity; police also found his searches about creating images with AI. The BBC court report says he used AI to make sexualised fake images from pupils' photographs and shared them online. The AI output is itself the sexual depiction of the pupils (depicted_or_impersonated). The tool is not named.","status":"supported"},"person_relations":["depicted_or_impersonated"]},"name":"Greater Manchester: former supply teacher jailed for six years; prosecutors say he used AI to make sexualised images of pupils from their social media photos","summary":"A former supply teacher who worked at secondary schools in Greater Manchester was jailed for six years at Bolton Crown Court on 15 October 2026 after pleading guilty to 21 offences, including publishing obscene material and taking indecent photographs of children. The Crown Prosecution Service says he used his position to find pupils' details, located their social media profiles and used AI to create sexualised images from their own photos, which he shared on social media; the BBC reports some were posted on X. Parents reported two accounts to police in March 2025. The court heard that one girl whose photograph had been used was left embarrassed. His offending also included more than 100 covert upskirt photographs taken in schools. The AI tool is not named.","incidentKind":"bounded_series","incidentDatePrecision":"unknown","exposurePattern":"unknown","reportedDate":"2026-10-15","aiSystem":"AI technology used to create sexualised images from pupils' social media photographs, as the Crown Prosecution Service describes it (tool not identified)","aiProduct":"Unidentified image tool","severity":"high","verificationStatus":"credible","harmCategories":[],"harmOutcomes":["exploitation_or_abuse","psychological_distress"],"harmOutcomeSummary":"According to the Crown Prosecution Service, a supply teacher used AI to make sexualised images of teenage pupils from their social media photos and shared them online; the court heard that one girl whose photo was used was left embarrassed.","frameworkFacets":[],"causationStatus":"supported","participantUsersAffectedMin":0,"otherPeopleHarmedMin":1,"affectedCountStatus":"partial","affectedCountEvidence":"At least one pupil whose social media photograph was used is described as embarrassed (claim c4, BBC court report). The CPS refers to victims in the plural and the court heard that 15 girls were targeted across all the offending, but the sources do not say how many were depicted in AI images, so further victims are unquantified. The pupils did not use the AI.","victimAgeRange":"minor","jurisdiction":"GB","platformType":"other","outcomeType":"criminal_charges","outcomeStatus":"resolved","primarySourceUrl":"https://www.bbc.co.uk/news/articles/ck79p174x3d3o","primarySourceLabel":"BBC News, 15 October 2026: Teacher jailed for taking upskirt photos of pupils","firstPublishedAt":"2026-10-16T03:00:14.536526+00:00","updatedAt":"2026-10-16T03:00:14.536526+00:00","scopeVersion":"facts-v3","tags":["ai-sexualised-images","non-consensual-sexual-imagery","minors","school-staff","social-media","guilty-plea","sentencing","england","united-kingdom","depicted-or-impersonated"]},{"id":"2025-mother-tells-lbc-chatgpt-reinforces-sons-delusions-during-manic-episodes","caseFacts":{"claims":[{"id":"c1","status":"reported","evidence":[{"locator":"'The mother-of-three, who has chosen to remain anonymous, said her son uses the chatbot daily and sees it as a reliable companion which can do no wrong.'; 'suffers from manic psychotic episodes.'","relation":"supports","source_id":"s1"}],"assertion":"A mother who asked to remain anonymous told LBC that her son, who has manic psychotic episodes, uses ChatGPT daily and sees it as a reliable companion that can do no wrong.","causal_attribution":"The mother's account to LBC; one origin."},{"id":"c2","status":"reported","evidence":[{"locator":"'He often turns to ChatGPT when these peak, asking it for advice about his grandiose ideas, she says.'","relation":"supports","source_id":"s1"}],"assertion":"She said he often turns to ChatGPT when his episodes peak, asking it for advice about his grandiose ideas.","causal_attribution":"The mother's account."},{"id":"c3","status":"reported","evidence":[{"locator":"'its more fanciful claims reinforce his florid delusions'; 'he is more willing to believe it that anything I have to say'","relation":"supports","source_id":"s1"}],"assertion":"She said the chatbot's more fanciful claims reinforce his florid delusions, and that he is more willing to believe it than anything she says.","causal_attribution":"The mother's attribution of the reinforcement to the chatbot; no chat log or clinical view was inspected."},{"id":"c4","status":"reported","evidence":[{"locator":"'This has left her feeling “anxious and defeated,” she says'; 'everything could come “crashing down”'","relation":"supports","source_id":"s1"}],"assertion":"She said the situation has left her feeling anxious and defeated, and that she fears everything could come crashing down because of his reliance on the technology.","causal_attribution":"The mother's own account of her distress."},{"id":"c5","status":"reported","evidence":[{"locator":"'People sometimes turn to ChatGPT in sensitive moments, so we want to make sure it responds appropriately, guided by experts.'; 'directing people to professional help when appropriate'","relation":"supports","source_id":"s1"}],"assertion":"In a statement published in the LBC article, an OpenAI spokesperson said that people sometimes turn to ChatGPT in sensitive moments and that the company wants it to respond appropriately, including by directing people to professional help.","causal_attribution":"The statement as published by LBC; it is general and does not address this family."}],"effects":[{"label":"delusions reinforced during manic psychotic episodes, by his mother's account","claim_id":"c3","direction":"negative"},{"label":"mother feels anxious and defeated","claim_id":"c4","direction":"negative"}],"sources":[{"id":"s1","url":"https://www.lbc.co.uk/news/exclusive/chatgpt-ai-psychosis-online-safety/","kind":"news_report","access":"read","language":"en","translation_note":"Read in English on 2026-10-15 (live page, HTTP 200; LBC, by Jacob Paul, dated 8 Sep 2025 and updated 10 Feb 2026). The case passages are the opening paragraphs quoting the mother. The rest of the article covers an academic's comments on emails he receives, UK legislation and statements from DSIT and OpenAI, which are not part of this case.","independence_group":"lbc"}],"version":1,"ai_roles":["own_use"],"contexts":["health","relationships"],"unknowns":["Any specific ChatGPT reply; LBC quotes none.","When the son's use and the episodes began, and whether he received clinical care.","The son's age and where the family lives.","The son's own view; only his mother was interviewed.","Any clinical assessment of the chatbot's role; none is reported."],"geography":{"basis":"LBC is a UK broadcaster, but the article does not say where the mother or son live, so no country is recorded.","court_countries":[],"event_countries":[],"affected_person_countries":[]},"publication":{"basis":"Published as a core case (communicated_with). The mother told LBC that her son asks ChatGPT about his grandiose ideas when his manic psychotic episodes peak and that its fanciful claims reinforce his delusions, and that this leaves her anxious and defeated. The account is hers alone and is attributed throughout; no reply, log or clinical view was inspected.","reviewed_on":"2026-10-15"},"ai_involvement":{"basis":"The mother told LBC that her son asks ChatGPT for advice about his grandiose ideas when his manic psychotic episodes peak, and that the chatbot's more fanciful claims reinforce his florid delusions. That describes what the chatbot said in his exchanges with it and connects it to the reinforcement of his delusions. LBC quotes no specific reply, and no chat logs, clinical records or statement from the son were inspected.","status":"reported"},"person_relations":["communicated_with"]},"name":"Mother tells LBC that ChatGPT's claims reinforce her son's delusions when his manic psychotic episodes peak, leaving her anxious","summary":"In a report published by LBC on 8 September 2025, a mother who asked to remain anonymous said her son, who has manic psychotic episodes, uses ChatGPT daily, sees it as a reliable companion that can do no wrong, and turns to it when the episodes peak to ask for advice about his grandiose ideas. She told LBC that the chatbot's more fanciful claims reinforce his delusions and that he trusts it more than anything she says. She said this has left her feeling anxious and defeated. The account is hers alone. LBC did not identify the son, and no chat logs, clinical records or statement from the son were inspected. The article includes a general statement from an OpenAI spokesperson about how ChatGPT responds in sensitive moments.","incidentKind":"ongoing_experience","incidentDatePrecision":"unknown","exposurePattern":"repeated_interactions","reportedDate":"2025-09-08","aiSystem":"ChatGPT (OpenAI chatbot)","aiProduct":"ChatGPT","aiCompany":"OpenAI","severity":"medium","verificationStatus":"unverified","harmCategories":["delusion_reinforcement"],"harmOutcomes":["psychiatric_deterioration","psychological_distress"],"harmOutcomeSummary":"By his mother's account to LBC, ChatGPT's claims reinforce her son's delusions when his manic psychotic episodes peak; she says the situation has left her feeling anxious and defeated.","frameworkFacets":[],"pathwayDirection":"inward","causationStatus":"alleged","participantUsersAffectedMin":1,"otherPeopleHarmedMin":1,"affectedCountStatus":"exact","affectedCountEvidence":"LBC describes one son who uses ChatGPT and whose delusions his mother says it reinforces, and the mother herself, who says she feels anxious and defeated. Her other children are not described as affected and are not counted.","victimAgeRange":"unknown","platformType":"chatbot","outcomeType":"media_coverage","outcomeStatus":"unknown","primarySourceUrl":"https://www.lbc.co.uk/news/exclusive/chatgpt-ai-psychosis-online-safety/","primarySourceLabel":"LBC, 8 September 2025: Worried mother says ChatGPT is fuelling son's delusions","firstPublishedAt":"2026-10-15T03:02:31.42524+00:00","updatedAt":"2026-10-15T03:02:31.42524+00:00","scopeVersion":"facts-v3","tags":["chatgpt","delusion","mania","family-account"]},{"id":"2026-united-airlines-chatbot-told-customer-travel-credit-lasted-five-years-airline-acknowledged-wrong-answer","caseFacts":{"claims":[{"id":"c1","status":"reported","evidence":[{"locator":"'she turned to a United Airlines chatbot for help understanding a $200 travel credit'; 'the chatbot told her the credit would remain active for five years'; 'stays active for 5 years from the date it'","relation":"supports","source_id":"s1"}],"assertion":"NBC 5 Chicago reports that a customer asked a United Airlines chatbot how long a $200 travel credit would last and that the chatbot said it would remain active for five years from deposit, as a screenshot the customer provided shows.","causal_attribution":"The customer's account and screenshot as described by the station; the airline later acknowledged the wrong answer (c4)."},{"id":"c2","status":"reported","evidence":[{"locator":"'a few weeks later, she received an email from United warning that her TravelBank cash would expire in a matter of months'; 'The email stated that her $200 credit was set to expire on September 27, 2026.'","relation":"supports","source_id":"s1"}],"assertion":"NBC 5 Chicago reports that a few weeks later United emailed the customer that the $200 credit was set to expire on 27 September 2026.","causal_attribution":"Reported by the station from the customer's email."},{"id":"c3","status":"reported","evidence":[{"locator":"'United apologized for the confusion and told her the travel credit could not be extended'; 'had hoped to save the credit for a larger trip'; 'she learned she had only months to use it before it expired'","relation":"supports","source_id":"s1"}],"assertion":"The customer says United at first apologised and said the credit lasted one year and could not be extended, so the customer, who had hoped to save it for a larger trip, had only months to use it.","causal_attribution":"The customer's account of the consequence."},{"id":"c4","status":"reported","evidence":[{"locator":"'AI chatbot provided this customer with the wrong information for her situation.'; 'we do provide a disclaimer informing customers that information may not always be complete or relevant'","relation":"supports","source_id":"s1"}],"assertion":"United told NBC 5 Responds that its AI chatbot provided the customer with the wrong information for the customer's situation, and that it shows a disclaimer that information may not always be complete or relevant.","causal_attribution":"The operator's own acknowledgement, quoted by the station."},{"id":"c5","status":"reported","evidence":[{"locator":"'The airline ultimately provided her with an electronic travel certificate.'; 'a $200 travel certificate for the inconvenience she experienced'","relation":"supports","source_id":"s1"}],"assertion":"After NBC 5 Responds contacted United, the airline sent the customer a $200 travel certificate.","causal_attribution":"Reported by the station."}],"effects":[{"label":"travel credit found to expire within months after the chatbot said it lasted five years","claim_id":"c3","direction":"negative"},{"label":"airline acknowledged the chatbot error and issued a $200 travel certificate after a media inquiry","claim_id":"c5","direction":"positive"}],"sources":[{"id":"s1","url":"https://www.nbcchicago.com/consumer/ai-chatbots-are-taking-over-customer-service-but-how-often-are-they-wrong/3989958/","kind":"news_report","access":"read","language":"en","translation_note":"Read in English on 2026-10-12 from the publisher page; no translation was needed and the research agent (an AI) read the text directly. The article quotes the customer, describes a screenshot the customer provided, and quotes a United statement given to the station.","independence_group":"nbc5-chicago-responds-united-chatbot"}],"version":1,"ai_roles":["institutional_use"],"contexts":["finance","everyday_life"],"unknowns":["When the chatbot exchange took place.","Where the customer lives.","Whether the customer would have lost the credit without the media inquiry.","Which policy applied to the funds; United says expiry varies with how funds were received."],"geography":{"basis":"The article, from a Chicago station, does not say where the customer lives or used the chatbot; no country is inferred from the station, the airline or the currency.","court_countries":[],"event_countries":[],"affected_person_countries":[]},"publication":{"basis":"Published as a concrete adverse experience with a customer-service chatbot: the chatbot's quoted answer about a travel credit was wrong, the operator acknowledged the error, and the customer faced an early expiry until the operator compensated after a media inquiry. Claims are attributed to the station, the customer and the airline.","reviewed_on":"2026-10-12"},"ai_involvement":{"basis":"The station describes a screenshot in which United's chatbot told the customer that the $200 TravelBank credit stays active for five years from deposit, and United itself told the station that its AI chatbot gave the customer wrong information. The report connects that answer to the customer finding the credit would expire within months and being told at first that it could not be extended.","status":"supported"},"person_relations":["communicated_with"]},"name":"United Airlines chatbot told a customer a $200 travel credit lasted five years; the airline later said it gave wrong information and issued a certificate","summary":"NBC 5 Chicago reported on 16 September 2026 that a customer asked a United Airlines chatbot how long a $200 TravelBank travel credit would remain usable and was told it stayed active for five years from deposit, as a screenshot the customer provided shows. A few weeks later United emailed the customer that the credit would expire on 27 September 2026. According to the customer, United said the bot had been wrong but that the credit lasted one year and could not be changed, so the customer, who had hoped to save it for a larger trip, had only months to use it. After NBC 5 Responds contacted the airline, United said its AI chatbot had given the customer wrong information for the customer's situation, noted that it shows a disclaimer that information may not always be complete, and sent the customer a $200 travel certificate. The report does not say when the chatbot exchange took place.","incidentKind":"single_event","incidentDatePrecision":"unknown","exposurePattern":"single_interaction","reportedDate":"2026-09-16","aiSystem":"United Airlines customer-service AI chatbot answering a question about a TravelBank travel credit","aiProduct":"United Airlines chatbot","severity":"low","verificationStatus":"credible","harmCategories":[],"harmOutcomes":["other_material_harm","psychological_distress"],"harmOutcomeSummary":"NBC 5 Chicago reports that a United Airlines chatbot told a customer a $200 travel credit lasted five years when it expired within months, leaving the customer facing an early loss of the credit and a dispute until the airline, after the station's inquiry, acknowledged the error and issued a $200 certificate.","frameworkFacets":[],"causationStatus":"supported","participantUsersAffectedMin":1,"otherPeopleHarmedMin":0,"affectedCountStatus":"exact","affectedCountEvidence":"One person: the customer who asked the chatbot about the travel credit and was told it lasted five years. Exact 1. The forum complaint the article mentions about another customer is not counted because it is not described beyond one sentence.","victimAgeRange":"unknown","jurisdiction":"unknown","platformType":"chatbot","outcomeType":"media_coverage","outcomeStatus":"resolved","primarySourceUrl":"https://www.nbcchicago.com/consumer/ai-chatbots-are-taking-over-customer-service-but-how-often-are-they-wrong/3989958/","primarySourceLabel":"NBC 5 Chicago, 16 September 2026: AI chatbots are taking over customer service. But how often are they wrong?","firstPublishedAt":"2026-10-12T03:22:22.061314+00:00","updatedAt":"2026-10-12T03:22:22.061314+00:00","scopeVersion":"facts-v3","tags":["customer-service-chatbot","airline","wrong-information","travel-credit","vendor-acknowledged"]},{"id":"2026-western-australia-man-jailed-ai-manipulated-images-of-real-children-exploitation-material","caseFacts":{"claims":[{"id":"c1","status":"reported","evidence":[{"locator":"'incorporating images of real victims and manipulating their images using AI tools'; 'was not mitigating because images of real children were used.'; 'used AI to manipulate images of children from social media.'","relation":"supports","source_id":"s1"}],"assertion":"The sentencing judge said the man was guilty of incorporating images of real victims and manipulating their images using AI tools, and that using AI to generate abusive material on the justification that it involved no actual abuse was not mitigating because images of real children were used; ABC reports he used AI to manipulate images of children from social media.","causal_attribution":"The judge's findings at sentencing after guilty pleas, as reported by one outlet; the sentencing remarks themselves were not inspected."},{"id":"c2","status":"reported","evidence":[{"locator":"'who had pleaded guilty to 26 charges including indecent recording, producing, creating and possessing child exploitation material.'; 'He also pleaded guilty to five charges of possession of indecent or obscene articles.'","relation":"supports","source_id":"s1"}],"assertion":"He pleaded guilty to 26 charges including indecent recording and producing, creating and possessing child exploitation material, and to five charges of possession of indecent or obscene articles.","causal_attribution":"Court report by ABC News."},{"id":"c3","status":"reported","evidence":[{"locator":"'He used encrypted messaging apps to send sexual material to co-accused'; 'as well as a covert police operative who was communicating with him.'; 'A police raid on his Wanneroo home yielded a large amount of child exploitation material'","relation":"supports","source_id":"s1"}],"assertion":"He used encrypted messaging apps to send sexual material to two co-accused and to a covert police operative, and a police raid on his Wanneroo home yielded a large amount of child exploitation material.","causal_attribution":"As heard in court and reported by ABC News; the report does not say which material sent was AI-manipulated."},{"id":"c4","status":"reported","evidence":[{"locator":"'was sentenced to four-and-a-half years in prison, and must serve two-and-a-half before being eligible for parole.'; 'He was fined $2,500 for the offensive material charges.'","relation":"supports","source_id":"s1"}],"assertion":"Judge Belinda Lonsdale of the WA District Court sentenced him to four and a half years in prison with eligibility for parole after two and a half years, and fined him $2,500 for the offensive material charges.","causal_attribution":"Court report by ABC News."}],"effects":[{"label":"images of real children were manipulated with AI tools into child exploitation material (sentencing judge as reported)","claim_id":"c1","direction":"negative"}],"sources":[{"id":"s1","url":"https://www.abc.net.au/news/2026-10-02/matthew-woodward-sentencing-ai-generated-child-exploitation/107216204","kind":"news_report","access":"read","language":"en","translation_note":"Read live on 2026-10-02 (ABC News, David Weber, posted 2 October 2026 at 7:16am AEST). Court reporting of the sentencing, quoting the judge and defence counsel. A photo caption in the article ('used AI to manipulate images of children from social media') is used alongside the body text.","independence_group":"abc-court-report"}],"version":1,"ai_roles":["others_use"],"contexts":["privacy","justice"],"unknowns":["When the offending took place and when the man was charged.","Which AI tools were used.","How many real children were depicted, whether any were known to him, and whether any were informed.","Which of the material sent to the co-accused and the police operative was AI-manipulated.","The sentencing date; the report was posted on 2 October 2026 (AEST)."],"geography":{"basis":"The man lived in Wanneroo, Western Australia, where police raided his home, and was sentenced in the WA District Court (ABC News). Where the depicted children live is not stated.","court_countries":["AU"],"event_countries":["AU"],"affected_person_countries":[]},"publication":{"basis":"Published as a core case (depicted_or_impersonated) consistent with existing AI child-abuse-image rows (for example the Colchester and Carter County rows): a sentencing judge, as reported by ABC News, found that a man manipulated images of real children with AI tools into child exploitation material, and he pleaded guilty and was jailed. One outlet reports the case, so claims are marked reported. Neither the man, the co-accused nor any child is named here.","reviewed_on":"2026-10-02"},"ai_involvement":{"basis":"ABC News reports the sentencing judge's finding that the man incorporated images of real victims and manipulated them using AI tools, and that images of real children were used; ABC also reports he used AI to manipulate images of children from social media. No tool is named.","status":"reported"},"person_relations":["depicted_or_impersonated"]},"name":"Western Australia: a 41-year-old Wanneroo man was jailed for four and a half years after pleading guilty to 26 charges including producing and possessing child exploitation material; the District Court judge said he had manipulated images of real victims with AI tools, and ABC News reports he used AI on images of children from social media","summary":"ABC News (2 October 2026) reports that WA District Court Judge Belinda Lonsdale sentenced a 41-year-old man to four and a half years in prison, with eligibility for parole after two and a half years, for creating child exploitation material with AI. He had pleaded guilty to 26 charges including indecent recording and producing, creating and possessing child exploitation material, and to five charges of possessing indecent or obscene articles, and was fined $2,500 for the offensive material charges. The judge said he was guilty of incorporating images of real victims and manipulating their images using AI tools, and that the claim that AI-generated material involved no actual abuse was not mitigating because images of real children were used; ABC reports he used AI to manipulate images of children from social media. He sent sexual material through encrypted messaging apps to two co-accused and to a covert police operative, and a police raid on his Wanneroo home found a large amount of child exploitation material. The children are not identified and their number is not reported.","incidentKind":"bounded_series","incidentDatePrecision":"unknown","exposurePattern":"unknown","reportedDate":"2026-10-02","aiSystem":"Unnamed AI tools used to manipulate images of real children, including images from social media, into child exploitation material (sentencing remarks as reported by ABC News)","aiProduct":"Unidentified image tool","severity":"high","verificationStatus":"credible","harmCategories":[],"harmOutcomes":["exploitation_or_abuse"],"harmOutcomeSummary":"Images of real children were manipulated with AI tools into child exploitation material, according to the sentencing judge as reported by ABC News, which also reports that he used AI on images of children from social media; the man pleaded guilty and was jailed for four and a half years.","frameworkFacets":[],"causationStatus":"supported","participantUsersAffectedMin":0,"otherPeopleHarmedMin":0,"affectedCountStatus":"unquantified","affectedCountEvidence":"The judge refers to images of real victims and real children manipulated with AI tools, and ABC reports images of children from social media, but no count of real children depicted is reported. Unquantified, numeric placeholders 0. Charge counts are not person counts.","victimAgeRange":"minor","jurisdiction":"AU-WA","platformType":"other","outcomeType":"criminal_charges","outcomeStatus":"resolved","primarySourceUrl":"https://www.abc.net.au/news/2026-10-02/matthew-woodward-sentencing-ai-generated-child-exploitation/107216204","primarySourceLabel":"ABC News, 2 October 2026: WA man who used AI to make child exploitation material jailed","firstPublishedAt":"2026-10-02T03:12:25.344122+00:00","updatedAt":"2026-10-02T03:12:25.344122+00:00","scopeVersion":"facts-v3","tags":["ai-csam","children","australia","western-australia","perth","sentencing","depicted-or-impersonated"]},{"id":"2025-montreal-icu-chief-francois-marquis-ai-deepfake-ads-scam-victim-icu","caseFacts":{"claims":[{"id":"c1","status":"reported","evidence":[{"locator":"'he said there has been a proliferation of artificial intelligence (AI) generated deepfakes on people’s social media feeds.'; 'It’s all people calling me and telling me there’s some deepfakes, you know, running around.'; 'he’s talking about joint pain and selling supplements. But more seriously, he said he’s seen videos where he is peddling cures for cancer and anti-pharmaceutical propaganda.'","relation":"supports","source_id":"s1"},{"locator":"'Marquis said he was in disbelief when he first learned his image was being used in a deepfake video.'; 'He received phone calls from people saying \"you're all over Facebook,\"'","relation":"supports","source_id":"s2"},{"locator":"'said he's heard from multiple patients and colleagues that videos featuring his likeness — all selling a range of pills or products — keep popping up on Facebook.'; 'referring to the latest AI-generated ad.'","relation":"supports","source_id":"s3"}],"assertion":"Dr. François Marquis, chief of intensive care at Montreal's Maisonneuve-Rosemont Hospital, says AI-generated deepfake videos using his likeness have circulated on Facebook since at least August 2025, selling joint supplements, pills and cancer cures and spreading anti-pharmaceutical claims; he learns of them from people who call him because he is not on social media.","causal_attribution":"The doctor's own account in three interviews (CBC 2025, CTV and CBC 2026); no inspected source reproduces or analyses a video."},{"id":"c2","status":"reported","evidence":[{"locator":"'This poor man actually ended up, you know, barging in the ICU and he wanted his money back because he never received the drugs,'","relation":"supports","source_id":"s1"},{"locator":"'after a person who was taken for a few hundred dollars in an online scam showed up at the hospital demanding his money back.'","relation":"supports","source_id":"s2"},{"locator":"'Marquis recalled a victim of one deepfake scam who came looking for him in the ICU, wanting his money back for pills that were never delivered.'; 'that person invested hundreds of dollars in those pills and never received anything,'","relation":"supports","source_id":"s3"}],"assertion":"A person who paid hundreds of dollars for pills advertised in one of the deepfakes, and never received them, came to the hospital and into the ICU demanding his money back from Dr. Marquis.","causal_attribution":"Recounted by Dr. Marquis in each interview; the person is not identified and did not speak to the outlets. The three accounts appear to describe the same visit, but no source confirms that."},{"id":"c3","status":"reported","evidence":[{"locator":"'has reported the deepfakes to Quebec’s College of Physicians and Montreal police, but he said there is little else he can do to make these videos stop, except speaking out about them.'","relation":"supports","source_id":"s1"},{"locator":"'he routinely struggles to get deepfake videos taken down before new ones pop back up — even though he's flagged them to social media platforms, Quebec's medical college and the police.'","relation":"supports","source_id":"s3"},{"locator":"'Basically I was told that there's nothing they can really do,'","relation":"supports","source_id":"s3"}],"assertion":"Dr. Marquis has reported the deepfakes to the Collège des médecins du Québec, Montreal police and social media platforms, but says new videos keep appearing and there is little he can do to stop them.","causal_attribution":"The doctor's account; he says he was told 'there's nothing they can really do', and no statement from the platforms, police or the college is reported."},{"id":"c4","status":"reported","evidence":[{"locator":"'now it's about the security of the people in the hospital,'; 'The other problem is that some people will actually stop taking their usual medication to take this fake drug,'","relation":"supports","source_id":"s2"}],"assertion":"He says the visit by the scam victim raised a security concern for people in the hospital, and that some people may stop their usual medication for fake drugs.","causal_attribution":"The doctor's stated concern; no patient who stopped treatment because of these videos is described."}],"effects":[{"label":"his face and voice used in AI-generated deepfake ads selling supplements, cancer cures and anti-pharmaceutical claims on Facebook","claim_id":"c1","direction":"negative"},{"label":"a person who paid hundreds of dollars for pills advertised in one of the deepfakes never received them and came to the ICU demanding his money back","claim_id":"c2","direction":"negative"},{"label":"reports to the medical college, police and platforms have not stopped the videos","claim_id":"c3","direction":"negative"}],"sources":[{"id":"s1","url":"https://www.ctvnews.ca/montreal/article/deepfakes-of-well-known-montreal-doctor-raising-alarm/","kind":"news_report","access":"read","language":"en","translation_note":"Read on 2026-10-01 from the live CTV News page (18 September 2026); article text from the page JSON-LD articleBody. Interview with Dr. Marquis.","independence_group":"marquis-own-account"},{"id":"s2","url":"https://www.cbc.ca/news/canada/montreal/quebec-doctors-deepfake-scams-warning-1.7599117","kind":"news_report","access":"read","language":"en","translation_note":"Read on 2026-10-01 from the live CBC News page (1 August 2025, updated 4 August 2025). Interview with Dr. Marquis and another Quebec physician.","independence_group":"marquis-own-account"},{"id":"s3","url":"https://ca.news.yahoo.com/deepfake-doctors-peddling-bogus-cures-080000588.html","kind":"news_report","access":"read","language":"en","translation_note":"Read on 2026-10-01 from the Yahoo News Canada copy of a CBC News feature (18 September 2026). Interview with Dr. Marquis and other physicians.","independence_group":"marquis-own-account"}],"version":1,"ai_roles":["others_use"],"contexts":["health","work","finance"],"unknowns":["Who made and paid for the ads, which AI tool was used, and how many videos exist.","Whether the ICU visit described in 2025 and in 2026 is the same event, and how many people bought products from the ads.","Whether Meta removed any of the reported ads, and the outcome of the police and medical-college reports.","When the first deepfake appeared (before August 2025)."],"geography":{"basis":"Dr. Marquis is chief of intensive care at Maisonneuve-Rosemont Hospital in Montreal, where the scam victim confronted him (CTV, CBC). The ads circulate on Facebook and their origin is not stated, so the event country is left unknown.","court_countries":[],"event_countries":[],"affected_person_countries":["CA"]},"publication":{"basis":"Published as a core case (depicted_or_impersonated): a named Montreal ICU chief, a public figure speaking on record to CBC and CTV, says AI deepfake ads using his likeness have sold supplements and cures on Facebook since at least August 2025, and that a scam victim who paid hundreds of dollars for undelivered pills confronted him in the ICU. The account rests on his own interviews; the ads' makers are unknown. The scam victim is not identified.","reviewed_on":"2026-10-01"},"ai_involvement":{"basis":"Dr. Marquis describes the videos as AI-generated deepfakes using his likeness (CTV News 2026; CBC 2025 and 2026), and CBC calls the latest one an AI-generated ad. No inspected source names the tool, the advertiser or who made the videos, and the doctor says he is not on social media and learns of the videos from people who call him.","status":"reported"},"person_relations":["depicted_or_impersonated"]},"name":"Montreal: Dr. François Marquis, chief of intensive care at Maisonneuve-Rosemont Hospital, says AI deepfake ads using his face have sold supplements and cancer cures on Facebook since at least August 2025, and that a scam victim who paid hundreds of dollars for pills that never arrived came to the ICU demanding a refund","summary":"Dr. François Marquis, chief of intensive care at Maisonneuve-Rosemont Hospital in Montreal, told CBC News (August 2025) and CTV News and CBC News (September 2026) that AI-generated deepfake videos using his likeness keep appearing on Facebook, selling joint supplements, pills and cancer cures and spreading anti-pharmaceutical claims; the latest offers $1 million to dissatisfied customers. He is not on social media and learns of the videos from people who call him. He recounts that a person who paid hundreds of dollars for pills advertised in one of the videos, and never received them, came into the ICU demanding his money back, which he describes as a security problem for the hospital. He has reported the deepfakes to Quebec's College of Physicians, Montreal police and the platforms, but says new videos keep appearing. The makers of the ads are not identified.","incidentKind":"ongoing_experience","incidentDatePrecision":"unknown","exposurePattern":"repeated_interactions","reportedDate":"2025-08-01","aiSystem":"AI-generated deepfake video ads using Dr. Marquis's face and voice, circulated on Facebook (his account to CBC and CTV); no tool or advertiser is named","aiProduct":"Unidentified video tool","severity":"low","verificationStatus":"credible","harmCategories":[],"harmOutcomes":["reputational_harm","financial_loss"],"harmOutcomeSummary":"Dr. Marquis says AI deepfake ads falsely show him selling supplements and cures and spreading anti-pharmaceutical claims, and that one person lost hundreds of dollars on pills that never arrived and confronted him in the ICU (his account to CBC and CTV).","frameworkFacets":[],"causationStatus":"alleged","participantUsersAffectedMin":0,"otherPeopleHarmedMin":2,"affectedCountStatus":"documented_minimum","affectedCountEvidence":"Dr. Marquis, whose likeness is used, and one person who paid hundreds of dollars for undelivered pills and came to the ICU (counted once; the 2025 and 2026 accounts appear to describe the same visit). The doctor says other people lose money but gives no number. Documented minimum 2.","victimAgeRange":"adult","jurisdiction":"CA-QC","platformType":"other","outcomeType":"media_coverage","outcomeStatus":"ongoing","primarySourceUrl":"https://www.ctvnews.ca/montreal/article/deepfakes-of-well-known-montreal-doctor-raising-alarm/","primarySourceLabel":"CTV News, 18 September 2026: Deepfakes of well-known Montreal doctor raising alarm","firstPublishedAt":"2026-10-01T03:18:04.682437+00:00","updatedAt":"2026-10-01T03:18:04.682437+00:00","scopeVersion":"facts-v3","tags":["deepfake","doctor-impersonation","health-scam","facebook-ads","montreal","quebec","depicted-or-impersonated"]},{"id":"2026-victoria-australia-ai-cloned-doctor-voice-gift-card-scam-mandarin-speakers","caseFacts":{"claims":[{"id":"c1","status":"reported","evidence":[{"locator":"'Scammers are using artificial intelligence to clone the voicers of doctors, general practitioners and surgeons.'; 'Ten reports of the scam were received last month'","relation":"supports","source_id":"s1"},{"locator":"'Police have received 10 reports since early September, with victims losing tens of thousands of dollars.'; 'The warning says the caller uses the victim's real name, then tells them to buy gift cards and hand over the gift card codes by text message.'; 'The scammers are using AI voice cloning and targeting Mandarin speakers across Victoria.'","relation":"supports","source_id":"s2"}],"assertion":"Victoria Police say scammers targeting Mandarin speakers are using artificial intelligence to clone the voices of doctors, general practitioners and surgeons, using victims' real names and telling them to buy gift cards and send the codes by text message; police received ten reports in September 2026.","causal_attribution":"Victoria Police warning carried by ABC and NewsWire (one chain)."},{"id":"c2","status":"reported","evidence":[{"locator":"'In one case, the scammer copied the voice of a medical professional known to the victim.'; 'The victim bought $50,000 in gift cards over three days, apparently as business gifts, until they became suspicious, called the medical professional's real number, and realised it was a scam.'","relation":"supports","source_id":"s2"},{"locator":"'one victim who lost $50,000 to the scam after buying gift cards across a three-day period before they became suspicious.'","relation":"supports","source_id":"s1"},{"locator":"'In one of the reports the scammer called a victim and mimicked the voice of a professional known to them.'; 'The victim purchased $50,000 in gift cards across a three-day period before becoming suspicious.'","relation":"supports","source_id":"s3"}],"assertion":"In one case the scammer copied the voice of a medical professional known to the victim, who bought $50,000 in gift cards over three days before becoming suspicious and calling the medical professional's real number.","causal_attribution":"Police account (one chain); the voice-copy detail is in the NewsWire copy only."},{"id":"c3","status":"reported","evidence":[{"locator":"'The other report saw the scammer pose as a medical professional and call the victim to exploit their urgent health conditions,'; 'The victim was repeatedly pressured into buying more cards until their cash flow was entirely depleted, resulting in a financial loss of $40,000.'","relation":"supports","source_id":"s2"},{"locator":"'Police say another victim lost $40,000 after a scammer pretended to be a medical professional who exploited their urgent health conditions.'","relation":"supports","source_id":"s1"}],"assertion":"In another case a scammer posing as a medical professional exploited the victim's urgent health conditions and repeatedly pressured them to buy gift cards until their cash flow was depleted, for a loss of $40,000.","causal_attribution":"Police account (one chain); police do not say whether a cloned voice was used in this case."}],"effects":[{"label":"one victim bought $50,000 in gift cards over three days after a caller used the copied voice of a medical professional the victim knew","claim_id":"c2","direction":"negative"},{"label":"another victim lost $40,000 after a caller posing as a medical professional exploited their urgent health conditions and pressured them until their cash was depleted","claim_id":"c3","direction":"negative"}],"sources":[{"id":"s1","url":"https://www.abc.net.au/news/2026-10-01/fake-doctor-scam-targets-mandarin-speaking-community/107214906","kind":"news_report","access":"read","language":"en","translation_note":"Read on 2026-10-01 from the live ABC News page (published 30 September 2026 UTC, dated 1 October locally); reports the Victoria Police warning and quotes the Cybercrime Squad.","independence_group":"victoria-police-warning"},{"id":"s2","url":"https://au.news.yahoo.com/aussies-targeted-chilling-ai-doctor-222212819.html","kind":"news_report","access":"read","language":"en","translation_note":"Read on 2026-10-01 from the live Yahoo News Australia page (NewsWire copy, 30 September 2026 UTC); quotes the Victoria Police warning at greater length.","independence_group":"victoria-police-warning"},{"id":"s3","url":"https://www.insidestategovernment.com.au/warning-issued-over-ai-fake-doctor-scam/","kind":"news_report","access":"read","language":"en","translation_note":"Read on 2026-10-01 from the live page (Inside State Government, 1 October 2026); reproduces the Victoria Police statement at greater length.","independence_group":"victoria-police-warning"}],"version":1,"ai_roles":["others_use"],"contexts":["health","finance"],"unknowns":["Which AI tool was used and how the voices were obtained.","How many of the ten reports involved a financial loss, and the total loss.","Whether a cloned voice was used in the $40,000 case.","The victims' ages and locations within Victoria, and the dates of the calls.","Whether the professional imitated in the $50,000 case was a medical professional: NewsWire says a medical professional and ABC says the victim called their health provider, while Inside State Government's account of the police statement says 'a professional known to them'.","How many of the ten reports involved a cloned voice."],"geography":{"basis":"Victoria Police say the calls have been reported across Victoria and are warning the Mandarin-speaking community in Victoria (ABC, Yahoo News Australia).","court_countries":[],"event_countries":["AU"],"affected_person_countries":["AU"]},"publication":{"basis":"Published as a core case (communicated_with): Victoria Police report a series of scam calls in which AI-cloned voices of doctors were used to make Mandarin-speaking victims buy gift cards, with two described victims losing $50,000 and $40,000. The AI element is the police account; no victim is identified.","reviewed_on":"2026-10-01"},"ai_involvement":{"basis":"Victoria Police, as reported by ABC News and Yahoo News Australia (NewsWire), say the scammers use artificial intelligence to clone the voices of doctors, general practitioners and surgeons; in the $50,000 case the scammer copied the voice of a medical professional known to the victim. Police do not say whether a cloned voice was used in the $40,000 case, and no tool is named. Inside State Government's account of the police statement says AI voice cloning 'has been featured in several reports to police' and describes the $50,000 case as the scammer having 'mimicked the voice of a professional known to them'.","status":"reported"},"person_relations":["communicated_with"]},"name":"Victoria, Australia: police warned, in reports published on 1 October 2026 local time, that scammers using AI to clone the voices of doctors have called Mandarin speakers and made them buy gift cards, with ten reports in September and two victims losing $50,000 and $40,000","summary":"Victoria Police warned, in reports published on 1 October 2026 local time, that fraudsters targeting the state's Mandarin-speaking community are using artificial intelligence to clone the voices of doctors, general practitioners and surgeons, calling victims by their real names and telling them to buy gift cards and text the codes (ABC News, Yahoo News Australia). Police received ten reports in September. In one case, NewsWire reports, the scammer copied the voice of a medical professional the victim knew, and the victim bought $50,000 in gift cards over three days before calling the real practice. In another, a caller posing as a medical professional exploited the victim's urgent health conditions and pressured them until their cash was depleted, a loss of $40,000; police do not say whether that call used a cloned voice. No victim is identified.","incidentKind":"bounded_series","incidentDatePrecision":"unknown","exposurePattern":"repeated_interactions","reportedDate":"2026-10-01","aiSystem":"AI voice cloning used by scammers to imitate the voices of doctors, general practitioners and surgeons in phone calls (Victoria Police, as reported); no tool is named","aiProduct":"Unidentified voice-cloning tool","severity":"medium","verificationStatus":"credible","harmCategories":[],"harmOutcomes":["financial_loss"],"harmOutcomeSummary":"Two victims described by Victoria Police lost $50,000 and $40,000 in gift cards after calls from scammers posing as doctors; police say the scammers clone doctors' voices with AI, and in the $50,000 case copied the voice of a medical professional the victim knew.","frameworkFacets":[],"causationStatus":"alleged","participantUsersAffectedMin":0,"otherPeopleHarmedMin":2,"affectedCountStatus":"documented_minimum","affectedCountEvidence":"Police describe two victims with losses ($50,000 and $40,000). Ten reports were received, but police do not say how many involved a loss. Documented minimum 2. Police do not say whether a cloned voice was used in the $40,000 case; both victims are counted as harmed within the reported scam series.","victimAgeRange":"unknown","jurisdiction":"AU-VIC","platformType":"other","outcomeType":"media_coverage","outcomeStatus":"unknown","primarySourceUrl":"https://www.abc.net.au/news/2026-10-01/fake-doctor-scam-targets-mandarin-speaking-community/107214906","primarySourceLabel":"ABC News, 1 October 2026: Scammers using AI to impersonate doctor in calls to Mandarin-speakers, police warn","firstPublishedAt":"2026-10-01T03:16:36.473962+00:00","updatedAt":"2026-10-01T03:16:36.473962+00:00","scopeVersion":"facts-v3","tags":["voice-cloning","impersonation-scam","gift-card-scam","mandarin-speakers","victoria","australia","communicated-with"]},{"id":"2026-80-year-old-woman-facebook-deepfake-romance-scam-140000-kidnap-ransom-story","caseFacts":{"claims":[{"id":"c1","status":"reported","evidence":[{"locator":"'In total she sent £140,000 to romance scammers.'; 'hollowed out a book and put £10,000 cash inside. She then posted it to a Shanghai Hospital where he awaited a life-saving operation.'; 'Then a new man came on the scene - he'd been kidnapped and if she didn't pay the ransom he would be killed.'; 'has returned £45,000 to her account.'","relation":"supports","source_id":"s1"}],"assertion":"An 80-year-old woman who believed she was in a relationship with a man she met on Facebook sent scammers £10,000 in cash hidden in a hollowed-out book, addressed to a Shanghai hospital where he supposedly awaited an operation; later 'a new man came on the scene' with a kidnap story and a ransom demand under threat of death; the BBC says she sent £140,000 in total, and her bank has returned £45,000.","causal_attribution":"Her daughter's account to the BBC; the BBC body does not itemise the payments or say whether a ransom was paid."},{"id":"c2","status":"reported","evidence":[{"locator":"'met her scammers on Facebook where they used artificial intelligence (AI) to change their appearance in photos and video calls - a type of deception known as deepfake.'; 'She says she quickly spotted the deepfake as the image was \"fuzzy\" and his voice out of sync with his movements.'","relation":"supports","source_id":"s1"}],"assertion":"The scammers used AI to change their appearance in photos and video calls; her daughter spotted the deepfake on a video call because the image was fuzzy and the voice out of sync.","causal_attribution":"The BBC's account, based on the daughter's interview."},{"id":"c3","status":"reported","evidence":[{"locator":"'tried to convince her mum the man she was speaking to was not who he said he was but she did not believe her.'; 'She was saying I made her feel disgusted and ashamed,'; 'She had a lasting power of attorney, meaning she could act on behalf of her mum and set up safeguards for her accounts.'; 'changed her mum's phone number three times, she continued to give the new one to the scammers.'; 'eventually replaced her mum's smartphone with one that could only do calls and texts.'","relation":"supports","source_id":"s1"}],"assertion":"She did not believe her daughter and said her daughter made her feel disgusted and ashamed; her daughter, who holds a lasting power of attorney, set up safeguards on her accounts, changed her phone number three times (each new number was given to the scammers) and eventually replaced her smartphone with one that could only make calls and send texts.","causal_attribution":"The daughter's account to the BBC."}],"effects":[{"label":"an 80-year-old woman sent scammers money, including £10,000 hidden in a book, and faced a ransom demand built on a kidnap story; the BBC says she sent £140,000 in total and her bank has returned £45,000","claim_id":"c1","direction":"negative"},{"label":"she rejected her daughter's attempts to intervene, saying her daughter made her feel disgusted and ashamed","claim_id":"c3","direction":"negative"}],"sources":[{"id":"s1","url":"https://www.bbc.com/news/articles/c674evx4p1x1o","kind":"news_report","access":"read","language":"en","translation_note":"Read on 2026-10-01 from the live BBC News page (published 29 September 2026 23:09 UTC); based on an interview with her daughter for BBC Radio 4's Scam Secrets. Brit Brief and other rewrites relay this report.","independence_group":"bbc-scam-secrets"}],"version":1,"ai_roles":["others_use"],"contexts":["relationships","finance"],"unknowns":["When the scam took place and over how long.","Where the woman lives.","How the £140,000 was paid, and whether a ransom was paid and how much.","Which AI tool was used, and whether voice was also generated.","Whether police were informed."],"geography":{"basis":"The BBC report (BBC Radio 4's Scam Secrets) does not state where the woman lives; it mentions UK fraud figures and a lasting power of attorney in general terms. Country left unknown.","court_countries":[],"event_countries":[],"affected_person_countries":[]},"publication":{"basis":"Published as a core case (communicated_with): the BBC reports, from her daughter's interview, that an 80-year-old woman was drawn into a Facebook romance scam in which the scammers used AI to alter their appearance in photos and video calls, and sent them money including £10,000 in a book, £140,000 in total according to the BBC; £45,000 has been returned. The scammers also told her of a kidnapping and demanded a ransom. Neither the woman nor her daughter is identified beyond the daughter's first name in the source.","reviewed_on":"2026-10-01"},"ai_involvement":{"basis":"The BBC reports that the scammers she met on Facebook used artificial intelligence to change their appearance in photos and video calls, and that her daughter, on a video call with the scammer, spotted the deepfake because the image was fuzzy and the voice out of sync with his movements. No tool is named.","status":"reported"},"person_relations":["communicated_with"]},"name":"An 80-year-old woman sent romance scammers she met on Facebook money including £10,000 hidden in a book after they used AI to alter their appearance in photos and video calls and later told her of a kidnapping with a ransom demand; the BBC reported (29 September 2026) that she sent £140,000 in total","summary":"The BBC (29 September 2026) reports, from an interview with her daughter for Radio 4's Scam Secrets, that an 80-year-old woman fell for a man she met on Facebook. The scammers used AI to change their appearance in photos and video calls. She posted £10,000 in cash hidden in a hollowed-out book to a Shanghai hospital where he supposedly awaited an operation; later, in the BBC's words, a new man came on the scene who had been kidnapped and would be killed if she did not pay a ransom. In total, the BBC says, she sent £140,000 to the scammers; her bank has returned £45,000. Her daughter spotted the deepfake on a video call, but the woman did not believe her and said her daughter made her feel disgusted and ashamed. The daughter, who holds a lasting power of attorney, set up safeguards with the bank, changed her mother's number three times (each new number reached the scammers) and eventually replaced her smartphone. The BBC does not say whether a ransom was paid.","incidentKind":"ongoing_experience","incidentDatePrecision":"unknown","exposurePattern":"cumulative_relationship","reportedDate":"2026-09-29","aiSystem":"AI used by romance scammers to change their appearance in photos and video calls (BBC); no tool is named","aiProduct":"Unidentified image and video tool","severity":"medium","verificationStatus":"credible","harmCategories":[],"harmOutcomes":["financial_loss","relationship_disruption"],"harmOutcomeSummary":"According to her daughter's account to the BBC, an 80-year-old woman sent romance scammers who used AI to alter their appearance money including £10,000 (£140,000 in total according to the BBC; £45,000 returned by her bank) and turned against her daughter's attempts to intervene.","frameworkFacets":[],"causationStatus":"alleged","participantUsersAffectedMin":0,"otherPeopleHarmedMin":1,"affectedCountStatus":"exact","affectedCountEvidence":"One person, the 80-year-old woman who sent the money (BBC). Her daughter's distress is reported but she is not counted. Exact 1.","victimAgeRange":"elderly","platformType":"other","outcomeType":"media_coverage","outcomeStatus":"unknown","primarySourceUrl":"https://www.bbc.com/news/articles/c674evx4p1x1o","primarySourceLabel":"BBC News, 29 September 2026: 'Mum sent them £140,000': How to spot the signs a loved one is being scammed","firstPublishedAt":"2026-10-01T03:16:29.04712+00:00","updatedAt":"2026-10-01T03:16:29.04712+00:00","scopeVersion":"facts-v3","tags":["romance-scam","deepfake","video-call","facebook","elderly","kidnap-ransom-story","communicated-with"]},{"id":"2025-sydney-school-reported-deepfake-images-police-inquiry","caseFacts":{"claims":[{"id":"c1","status":"reported","evidence":[{"locator":"Article body, retained text line 99: paragraph about digitally altered images.","relation":"supports","source_id":"s1"}],"assertion":"ABC reports that digitally altered sexual images using female students' faces circulated online at a Sydney high school.","causal_attribution":"Two ABC reports on one school investigation. Police and commissioner statements remain reported. General statistics or experiences at other schools do not establish this case's individual impacts."},{"id":"c2","status":"reported","evidence":[{"locator":"Article body, retained text line 93: paragraph about police complaint and investigation.","relation":"supports","source_id":"s1"},{"locator":"Article body, retained text line 101: paragraph about police complaint and investigation.","relation":"supports","source_id":"s1"}],"assertion":"Families reportedly went to police and NSW Police confirmed an investigation was underway.","causal_attribution":"Two ABC reports on one school investigation. Police and commissioner statements remain reported. General statistics or experiences at other schools do not establish this case's individual impacts."},{"id":"c3","status":"reported","evidence":[{"locator":"Article body, retained text line 99: paragraph about eSafety involvement.","relation":"supports","source_id":"s2"}],"assertion":"The follow-up says the eSafety Commissioner's investigators were in talks with police and the education department about the digitally altered images.","causal_attribution":"Two ABC reports on one school investigation. Police and commissioner statements remain reported. General statistics or experiences at other schools do not establish this case's individual impacts."}],"effects":[{"label":"Reported nonconsensual sexual depictions circulated online","claim_id":"c1","direction":"negative"}],"sources":[{"id":"s1","url":"https://www.abc.net.au/news/2025-10-16/police-investigate-sydney-school-image-deepfake/105896812","kind":"journalistic","access":"read","language":"en","translation_note":"","independence_group":"original-account-aiid-1234"},{"id":"s2","url":"https://www.abc.net.au/news/2025-10-17/deepfake-image-based-abuse-doubles-across-australia/105905152","kind":"journalistic","access":"read","language":"en","translation_note":"","independence_group":"original-account-aiid-1234"}],"version":1,"ai_roles":["others_use"],"contexts":["education","privacy"],"unknowns":["The first creation date, generation method and number of affected students are unknown.","The identities of creators and any investigation or disciplinary outcome remain unknown."],"geography":{"basis":"Both reports explicitly locate the affected students at a Sydney high school. The school's name is withheld.","court_countries":[],"event_countries":["AU"],"affected_person_countries":["AU"]},"publication":{"basis":"Inspected local reporting describes a concrete school image-abuse event and case-specific official responses. Students and school are not named. The record remains undated and does not borrow distress claims from other schools or general statistics.","reviewed_on":"2026-09-30"},"ai_involvement":{"basis":"ABC describes the case as deepfake image-based abuse and reports the eSafety Commissioner's involvement. The reviewers have not inspected the images, tools or a forensic determination.","status":"reported"},"person_relations":["depicted_or_impersonated"]},"name":"Sydney schoolgirls reportedly targeted with deepfake images circulated online","summary":"ABC News reports that digitally altered sexual images using the faces of girls from a Sydney high school circulated online. Families went to police, who confirmed an investigation, while the school and education department said they were working with police. A follow-up report says the eSafety Commissioner was also in talks with investigators. The school is not identified, and the number of affected students, creation date and later inquiry outcome are unknown.","incidentKind":"single_event","incidentDatePrecision":"unknown","exposurePattern":"unknown","reportedDate":"2025-10-16","aiSystem":"Reported deepfake image-generation tools","aiProduct":"Unidentified image tool","aiCompany":"Unknown","severity":"medium","verificationStatus":"credible","harmCategories":[],"harmOutcomes":["exploitation_or_abuse"],"harmOutcomeSummary":"The reports describe nonconsensual sexual depictions of schoolgirls circulated online and families seeking police help. No separate clinical or educational outcome is attributed to these particular students.","frameworkFacets":[],"causationStatus":"alleged","participantUsersAffectedMin":0,"otherPeopleHarmedMin":0,"affectedCountStatus":"unquantified","affectedCountEvidence":"The sources refer to female students and several families without a reliable numeric count. Zero is an unknown-count placeholder and does not mean no one was harmed.","victimAgeRange":"minor","platformType":"other","primarySourceUrl":"https://www.abc.net.au/news/2025-10-16/police-investigate-sydney-school-image-deepfake/105896812","primarySourceLabel":"ABC News, 16 October 2025, police inquiry into Sydney school deepfake images","firstPublishedAt":"2026-09-30T01:10:39.838754+00:00","updatedAt":"2026-09-30T01:58:17.276886+00:00","scopeVersion":"facts-v3","tags":[]},{"id":"2025-chicago-reported-musk-voice-impersonation-crypto-loss","caseFacts":{"claims":[{"id":"c1","status":"reported","evidence":[{"locator":"Article body, retained text line 20: paragraph about AI voice memos.","relation":"supports","source_id":"s1"}],"assertion":"ABC7 reports that the scammer sent the man fake audio messages using Elon Musk's voice, assembled with artificial intelligence.","causal_attribution":"One ABC7 interview and the outlet's report about the audio. The loss and consequences are the interviewee's account, without proof that AI alone caused the transfer."},{"id":"c2","status":"reported","evidence":[{"locator":"Article body, retained text line 15: paragraph about cryptocurrency transfers and discovery.","relation":"supports","source_id":"s1"},{"locator":"Article body, retained text line 15: paragraph about cryptocurrency transfers and discovery.","relation":"supports","source_id":"s1"}],"assertion":"The man said he sent a total of $10,000 in cryptocurrency and eventually realised it was a scam when he could not recover the money.","causal_attribution":"One ABC7 interview and the outlet's report about the audio. The loss and consequences are the interviewee's account, without proof that AI alone caused the transfer."},{"id":"c3","status":"reported","evidence":[{"locator":"Article body, retained text line 38: paragraph about difficulty paying expenses.","relation":"supports","source_id":"s1"}],"assertion":"He said the loss left him struggling to pay bills, rent and car expenses.","causal_attribution":"One ABC7 interview and the outlet's report about the audio. The loss and consequences are the interviewee's account, without proof that AI alone caused the transfer."}],"effects":[{"label":"Reported $10,000 cryptocurrency loss","claim_id":"c2","direction":"negative"}],"sources":[{"id":"s1","url":"https://abc7chicago.com/post/celebrity-crypto-cons-chicago-area-man-richard-lyons-loses-10k-cryptocurrency-scam-elon-musk-impersonator/17007708/","kind":"journalistic","access":"read","language":"en","translation_note":"","independence_group":"original-account-aiid-1142"}],"version":1,"ai_roles":["others_use"],"contexts":["finance","everyday_life"],"unknowns":["The first event date, scammer identity and audio-generation tool are unknown.","Whether any funds were subsequently recovered is unknown."],"geography":{"basis":"ABC7 identifies the interviewee as a Chicago-area resident and places the experience in that setting.","court_countries":[],"event_countries":["US"],"affected_person_countries":["US"]},"publication":{"basis":"A direct interview links a concrete financial loss to reported AI voice impersonation. The man's identity is minimized. The event date remains unknown and July 2025 is only the report date.","reviewed_on":"2026-09-30"},"ai_involvement":{"basis":"ABC7 specifically reports fake voice memos using Musk's voice assembled with artificial intelligence in this man's contact with the scammer. This is a journalistic attribution, without inspected forensic analysis.","status":"reported"},"person_relations":["communicated_with"]},"name":"Chicago-area man reports $10,000 loss after AI voice messages impersonating Elon Musk","summary":"A Chicago-area man told ABC7 that he sent $10,000 in cryptocurrency to someone posing as Elon Musk. The outlet reports that the scammer used AI-assembled voice messages using Musk's voice, alongside social media messages, apparent investment returns and a promised vehicle. The man said he could not retrieve the money after several months and was struggling to pay ordinary bills. The beginning of the contact and the technical creation of the messages remain unknown.","incidentKind":"single_event","incidentDatePrecision":"unknown","exposurePattern":"unknown","reportedDate":"2025-07-08","aiSystem":"Reported AI-generated voice impersonation","aiProduct":"Unidentified voice-cloning tool","aiCompany":"Unknown","severity":"medium","verificationStatus":"credible","harmCategories":[],"harmOutcomes":["financial_loss"],"harmOutcomeSummary":"The interviewee reported losing $10,000 in cryptocurrency and difficulty paying rent, bills and car expenses.","frameworkFacets":[],"causationStatus":"alleged","participantUsersAffectedMin":0,"otherPeopleHarmedMin":1,"affectedCountStatus":"documented_minimum","affectedCountEvidence":"One interviewed man describes his own loss. His relatives and the impersonated actor are not counted.","victimAgeRange":"adult","platformType":"other","primarySourceUrl":"https://abc7chicago.com/post/celebrity-crypto-cons-chicago-area-man-richard-lyons-loses-10k-cryptocurrency-scam-elon-musk-impersonator/17007708/","primarySourceLabel":"ABC7 Chicago I-Team, 8 July 2025, interview with scam victim","firstPublishedAt":"2026-09-30T01:10:26.782051+00:00","updatedAt":"2026-09-30T01:57:44.387352+00:00","scopeVersion":"facts-v3","tags":[]},{"id":"2026-tasmania-school-reported-ai-sexual-images-of-girls-shared-in-boys-group-chat","caseFacts":{"claims":[{"id":"c1","status":"reported","evidence":[{"locator":"have had their images used to make deepfake pornographic material, according to Tasmania Police.","relation":"supports","source_id":"s1"},{"locator":"who told them the AI-altered photos of the girls had been shared in a boys' group chat and that their images appeared to have been taken from social media.","relation":"supports","source_id":"s1"},{"locator":"The faked images of the girls were allegedly shared in a group chat by male students.","relation":"supports","source_id":"s2"},{"locator":"It is understood the incident may have involved the alleged use of AI-generated content.","relation":"context","source_id":"s3"}],"assertion":"Tasmania Police said, as reported by the ABC, that a number of girls at the school had their likenesses used to make AI-generated sexual images. Two parents said police told them the AI-altered photos were shared in a boys' group chat and appeared to have been taken from social media.","causal_attribution":"A Tasmania Police attribution stated by the ABC and two parents' account of what police told them. The images and the AI tool were not verified by the reviewer and no tool is named."},{"id":"c2","status":"reported","evidence":[{"locator":"Parents of two of the children said they were told by police that 21 victims had been identified.","relation":"supports","source_id":"s1"},{"locator":"The parents said police told them 21 girls had been identified as victims.","relation":"supports","source_id":"s2"}],"assertion":"Two parents said police told them 21 girls had been identified as victims.","causal_attribution":"Second-hand: two parents relaying a police figure to one ABC reporter. Police published no count in the inspected sources."},{"id":"c3","status":"reported","evidence":[{"locator":"Five boys are alleged to have created the images and all have since left the school.","relation":"supports","source_id":"s1"},{"locator":"Five boys have left the school and police said no charges had been laid.","relation":"supports","source_id":"s2"}],"assertion":"Five boys are alleged to have created the images and have left the school.","causal_attribution":"An allegation the ABC states in the sentence after the parents' account of what police told them, without naming its source. No charge or finding establishes who created the images."},{"id":"c4","status":"reported","evidence":[{"locator":"Both parents said their daughters were struggling with not being able to talk about it with their friends.","relation":"supports","source_id":"s1"},{"locator":"This has created an environment of confusion, isolation, and shame amongst the very girls who need support and solidarity the most.","relation":"supports","source_id":"s1"},{"locator":"The two parents told the ABC their daughters felt isolated and unsupported because they did not know if any of their peers had been told.","relation":"supports","source_id":"s2"},{"locator":"not knowing who to talk to","relation":"supports","source_id":"s1"}],"assertion":"Two parents said their daughters felt isolated and unsupported because they did not know whether peers had been told. One mother told the education minister in a letter that the girls' uncertainty about what they could say to students who had not been told had created confusion, isolation and shame among the girls.","causal_attribution":"Parents' accounts relayed by the ABC. No girl is quoted and no clinical assessment is reported."},{"id":"c5","status":"disputed","evidence":[{"locator":"the staff member told her most parents were not going to tell their children they had been identified in the material","relation":"supports","source_id":"s1"},{"locator":"I felt I was being encouraged not to tell my daughter","relation":"supports","source_id":"s1"},{"locator":"that most parents were not going to tell their daughters","relation":"supports","source_id":"s1"},{"locator":"and went on to advise them that most parents were not going to tell their children","relation":"supports","source_id":"s2"},{"locator":"At no time have families been advised or encouraged to withhold information from their child, nor to prevent their child from speaking with trusted adults.","relation":"contradicts","source_id":"s2"}],"assertion":"Two parents said senior school staff told them most parents were not going to tell their children they had been identified. The school principal wrote to families that at no time had families been advised or encouraged to withhold information.","causal_attribution":"A parents' account and the principal's written denial, both relayed by the ABC. The dispute concerns the school's handling, not how the images were made."},{"id":"c6","status":"reported","evidence":[{"locator":"Tasmania Police has received a report regarding alleged illegal online activity involving a group of students at a southern Tasmanian school","relation":"supports","source_id":"s3"},{"locator":"It is understood the incident may have involved the alleged use of AI-generated content.","relation":"supports","source_id":"s3"}],"assertion":"Tasmania Police received a report of alleged illegal online activity involving a group of students at a school in the south of the state. On 19 March 2026 Pulse Tasmania reported that, as the outlet understood it, the incident may have involved AI-generated content.","causal_attribution":"A police spokesperson statement to Pulse Tasmania. The AI involvement is the outlet's understanding and is not attributed to police in this article."},{"id":"c7","status":"reported","evidence":[{"locator":"Police said they have spoken to five young people and their parents following concerns about inappropriate online activity.","relation":"supports","source_id":"s4"},{"locator":"The youths were dealt with under the Youth Justice Act and no charges were laid.","relation":"supports","source_id":"s4"},{"locator":"Tasmania Police said no charges had been laid, with \"the five youths being dealt with in accordance with the Youth Justice Act\".","relation":"supports","source_id":"s1"}],"assertion":"Police spoke to five young people and their parents. No charges were laid and the five youths were dealt with under the Youth Justice Act.","causal_attribution":"Tasmania Police statements given to Pulse Tasmania (27 March 2026) and the ABC (29 April 2026). Both may relay one police position, so no independent corroboration is claimed."}],"effects":[{"label":"Girls reportedly depicted in AI-generated sexual images shared in a boys' group chat","claim_id":"c1","direction":"negative"},{"label":"Parents report the girls felt isolated, unsupported and unsure whom to talk to","claim_id":"c4","direction":"negative"}],"sources":[{"id":"s1","url":"https://www.abc.net.au/news/2026-04-29/tas-friends-school-criticised-after-deepfake-scandal/106577310","kind":"news_report","access":"read","language":"en","translation_note":"Full article body read (29 April 2026, updated 30 April). Related-story cards on the page were not used.","independence_group":"abc-tasmania-reporting-2026-04"},{"id":"s2","url":"https://www.abc.net.au/news/2026-04-30/tas-friends-school-deny-dont-tell-to-parents-over-deepfakes/106624754","kind":"news_report","access":"read","language":"en","translation_note":"Full article body read (30 April 2026, updated 1 May). Same reporter and same two parents as s1, so one independence group.","independence_group":"abc-tasmania-reporting-2026-04"},{"id":"s3","url":"https://pulsetasmania.com.au/news/online-activity-of-friends-school-students-under-police-investigation/","kind":"news_report","access":"read","language":"en","translation_note":"Full article body read. Based on a police spokesperson statement and a school letter to parents. Publication date 19 March 2026 from the article header line in the saved text ('Thu March 19, 2026 5:00am'), which matches the published-time metadata in the saved html.","independence_group":"pulse-tasmania-reporting-2026-03"},{"id":"s4","url":"https://pulsetasmania.com.au/news/no-charges-laid-after-friends-school-students-alleged-online-behaviour/","kind":"news_report","access":"read","language":"en","translation_note":"Full article body read. Same outlet and same police and school chain as s3. Publication date 27 March 2026 from the article header line in the saved text ('Fri March 27, 2026 2:52pm'), which matches the published-time metadata in the saved html.","independence_group":"pulse-tasmania-reporting-2026-03"}],"version":1,"ai_roles":["others_use"],"contexts":["education","privacy"],"unknowns":["No source states when the images were created or first shared. The earliest date in the sources is early March 2026, when a parent said the incident was being talked about. An ABC item of 5 July 2026 describes the incident as occurring 'earlier this year', and an ABC podcast page of 8 May 2026 asks what teenagers were doing 'over the school holidays' and says the images 'might not have been created in the classroom'. Neither gives a date, so the event start date is unknown.","The AI tool and the app or platform used for the group chat are not reported.","The girls' ages and school years are not reported. The sources call them girls, students, daughters and children.","The count of 21 comes from two parents relaying what police told them. Police published no count in the inspected sources, and the number of girls not yet told is unknown.","A local outlet quotes a school letter referring to students from the school and from other local schools. Whether students from other schools created, received or were depicted in the images is not reported.","How each girl was affected beyond the parents' accounts is not reported, and no girl is quoted.","The identity of the five youths and any later police or school outcome beyond the dealt-with-under-the-Youth-Justice-Act statement are not reported."],"geography":{"basis":"The sources describe a school in Tasmania, Tasmania Police, a police statement citing the Youth Justice Act and the Tasmanian education minister. No court proceeding is reported.","court_countries":[],"event_countries":["AU"],"affected_person_countries":["AU"]},"publication":{"basis":"Reports by the ABC (29 and 30 April 2026) and Pulse Tasmania (19 and 27 March 2026). All claims are reported except one that is disputed (the school's advice to parents): the victim count, the alleged creators and the images come from parents relaying police information and from police statements to the two outlets. The school, students, parents, principal and staff are not named and no image detail is given. No source dates the creation of the images, so the case is published undated and outside the 2026 event-year lane.","reviewed_on":"2026-09-30"},"ai_involvement":{"basis":"Tasmania Police, as relayed by the ABC, said girls' likenesses were used to make AI-generated sexual images, and two parents said police described the photos as AI-altered. The first local report (19 March 2026) said only that the incident may have involved AI-generated content. No AI tool is named and the images were not seen by the reviewer.","status":"reported"},"person_relations":["depicted_or_impersonated"]},"name":"Girls at a Tasmanian school reportedly depicted in AI-generated sexual images shared in a boys' group chat","summary":"Police in Tasmania received a report about alleged illegal online activity involving a group of students at a school in the south of the state, and on 19 March 2026 a local outlet reported that the incident may have involved AI-generated content. On 29 April 2026 the ABC reported, citing Tasmania Police, that girls at the school had their likenesses used to make AI-generated sexual images. Two parents told the ABC that police said 21 girls had been identified as victims, that the AI-altered photos were shared in a boys' group chat and that the images appeared to have been taken from social media. Five boys are alleged to have created the images and have left the school. Tasmania Police said no charges were laid and the five youths were dealt with under the Youth Justice Act. The parents said their daughters felt isolated and unsupported, and the school principal denied that families were advised to withhold information from their children. The creation date and the AI tool are not reported.","incidentKind":"single_event","incidentDatePrecision":"unknown","exposurePattern":"unknown","reportedDate":"2026-03-19","aiSystem":"Unspecified AI image-manipulation software","aiProduct":"Unidentified image tool","aiCompany":"Unknown","severity":"medium","verificationStatus":"credible","harmCategories":[],"harmOutcomes":["exploitation_or_abuse","psychological_distress"],"harmOutcomeSummary":"Tasmania Police, as relayed by the ABC, said girls' likenesses were used to make AI-generated sexual images, and two parents said police told them 21 girls had been identified as victims. The parents said their daughters felt isolated, unsupported and unsure whom to talk to, and a mother's letter to the education minister describes confusion, isolation and shame among the girls. These are parents' accounts relayed by one outlet and no girl is quoted.","frameworkFacets":[],"causationStatus":"alleged","participantUsersAffectedMin":0,"otherPeopleHarmedMin":21,"affectedCountStatus":"documented_minimum","affectedCountEvidence":"Two parents told the ABC that police told them 21 girls had been identified as victims (ABC, 29 and 30 April 2026). The figure is a police count relayed by two parents to one reporter and repeated by the outlet. Police published no count in the inspected sources. 21 is used as a lower bound because it counts identified victims only.","victimAgeRange":"minor","platformType":"other","primarySourceUrl":"https://www.abc.net.au/news/2026-04-29/tas-friends-school-criticised-after-deepfake-scandal/106577310","primarySourceLabel":"ABC News (29 Apr 2026): parents criticise a Tasmanian school response after girls were targeted in an AI deepfake incident","firstPublishedAt":"2026-09-30T01:09:53.76606+00:00","updatedAt":"2026-09-30T01:44:24.955277+00:00","scopeVersion":"facts-v3","tags":[]},{"id":"2025-peru-mollepata-tourist-reportedly-paid-160-dollars-sacred-canyon-of-humantay-operator-says-nonexistent","caseFacts":{"claims":[{"id":"c1","status":"reported","evidence":[{"locator":"tourists were chatting amicably about their plans to hike alone in the mountains to the \"Sacred Canyon of Humantay\".","relation":"supports","source_id":"s1"},{"locator":"They [showed] me the screenshot, confidently written and full of vivid adjectives","relation":"supports","source_id":"s1"},{"locator":"it was not true. There is no Sacred Canyon of Humantay!","relation":"supports","source_id":"s1"},{"locator":"O Cânion Sagrado de Humantay não existe!","relation":"supports","source_id":"s2"}],"assertion":"A trek operator says two tourists showed the operator a screenshot describing a 'Sacred Canyon of Humantay' and that no such place exists.","causal_attribution":"The operator's statement. The BBC does not verify the screenshot or name the AI tool. An academic quoted by the BBC believes the program 'likely just put together a few words that seemed approriate to the region'."},{"id":"c2","status":"reported","evidence":[{"locator":"The tourist paid nearly $160 (£118) in order to get to a rural road in the environs of Mollepata without a guide or [a destination].","relation":"supports","source_id":"s1"},{"locator":"Os turistas pagaram cerca de US$ 160","relation":"context","source_id":"s2"}],"assertion":"According to the operator, one tourist paid nearly $160 to reach a rural road near Mollepata without a guide or a destination.","causal_attribution":"The operator's statement. The BBC's account of the sequence (payment before or after the operator met the tourists) is ambiguous."},{"id":"c3","status":"reported","evidence":[{"locator":"the AI program likely just put together a few words that seemed approriate to the region.","relation":"supports","source_id":"s1"}],"assertion":"The BBC attributes the invented canyon to an AI program, quoting an academic's belief that the program likely combined words that seemed appropriate to the region.","causal_attribution":"The academic's belief as quoted by the BBC. No tool or chat was seen."}],"effects":[{"label":"Reported wasted fare and a trip to a rural road with no guide and no real destination","claim_id":"c2","direction":"negative"}],"sources":[{"id":"s1","url":"https://www.bbc.com/travel/article/20250926-the-perils-of-letting-ai-plan-your-next-trip","kind":"news_report","access":"read","language":"en","translation_note":"English original, no translation.","independence_group":"bbc-travel-ai-trip-planning-2025-09-29"},{"id":"s2","url":"https://www1.folha.uol.com.br/tec/2025/10/os-perigos-de-deixar-que-a-ia-organize-sua-proxima-viagem.shtml","kind":"news_report","access":"read","language":"pt","translation_note":"Portuguese translation of the BBC article published in Folha. Read by an AI model without human translation review. Not independent of s1.","independence_group":"bbc-travel-ai-trip-planning-2025-09-29"}],"version":1,"ai_roles":["own_use"],"contexts":["everyday_life"],"unknowns":["Date of the event is not stated in any inspected source.","The AI tool the tourists used is not named for this event, and the screenshot is not shown.","The tourists' identities, nationalities and what happened after the operator's intervention are not reported.","The BBC text refers to 'the tourist' paying although it introduces two tourists. Whether both paid is unknown.","The BBC does not say to whom the fare was paid, whether it was paid before or after the operator spoke to the tourists, or whether it was lost.","The operator's own company page tells a differing story of a German couple who were re-planned onto a real route. That page names no payment and says the couple 'still hiked'. It is not established whether that is the same event.","The BBC article is a trend feature. Only the operator's account is given, from a single witness."],"geography":{"basis":"BBC places the operator in a rural Peruvian town and the road near Mollepata. The tourists' nationality is not stated in the BBC article.","court_countries":[],"event_countries":["PE"],"affected_person_countries":[]},"publication":{"basis":"A named local trek operator's account to the BBC, relayed in translation by Folha. One source chain. Unverified, undated and the AI tool is not identified, so the event date is not established and the case is published undated.","reviewed_on":"2026-09-29"},"ai_involvement":{"basis":"The operator says the tourists showed the operator a screenshot describing the canyon. The BBC ties the invention to AI chatbots and quotes an expert on the likely origin, but does not state which tool the tourists used or show the screenshot.","status":"reported"},"person_relations":["communicated_with"]},"name":"Tourist reportedly paid nearly $160 to reach a rural road near Mollepata for a 'Sacred Canyon of Humantay' that a trek operator says does not exist","summary":"BBC Travel reported on 29 September 2025 that a trek operator in a rural Peruvian town overheard two unaccompanied tourists planning to hike alone to the 'Sacred Canyon of Humantay'. The operator says the tourists showed the operator a screenshot of a confident, adjectival description, that no such canyon exists, and that one tourist paid nearly $160 to reach a rural road near Mollepata without a guide or a destination. The article implies an AI chatbot as the source of the description but does not name the tool used by these tourists, and gives no date.","incidentKind":"single_event","incidentDatePrecision":"unknown","exposurePattern":"single_interaction","reportedDate":"2025-09-29","aiSystem":"Unidentified AI chatbot","aiProduct":"Unidentified chatbot","severity":"low","verificationStatus":"unverified","harmCategories":[],"harmOutcomes":["other_material_harm"],"harmOutcomeSummary":"According to the trek operator quoted by the BBC, one tourist paid nearly $160 to travel to a rural road with no guide and no real destination. The operator says the plan could have exposed them to high-altitude risk. No injury is reported.","frameworkFacets":[],"causationStatus":"alleged","participantUsersAffectedMin":1,"otherPeopleHarmedMin":0,"affectedCountStatus":"documented_minimum","affectedCountEvidence":"The BBC reports that 'the tourist' paid the fare and ended on a rural road. A second tourist is mentioned as part of the pair but no loss is reported for that person.","victimAgeRange":"adult","platformType":"chatbot","primarySourceUrl":"https://www.bbc.com/travel/article/20250926-the-perils-of-letting-ai-plan-your-next-trip","primarySourceLabel":"BBC Travel, \"The perils of letting AI plan your next trip\", 29 Sep 2025","firstPublishedAt":"2026-09-29T13:29:24.386049+00:00","updatedAt":"2026-09-30T01:17:13.239892+00:00","scopeVersion":"facts-v3","tags":[]},{"id":"2026-queensland-school-teacher-charged-over-alleged-ai-images-of-staff-and-students","caseFacts":{"claims":[{"id":"c1","status":"reported","evidence":[{"locator":"Police allege the man used AI software to generate explicit images of staff and students.","relation":"supports","source_id":"s1"},{"locator":"used a work-issued laptop and AI software to generate explicit images depicting staff members and students.","relation":"supports","source_id":"s1"},{"locator":"the AI content was found in a former secondary teacher's school digital account.","relation":"supports","source_id":"s2"}],"assertion":"Police allege that a former secondary school teacher used AI software on a work-issued laptop to generate explicit images depicting staff members and students of the school.","causal_attribution":"The allegation comes from a police statement and a school letter to parents, both relayed by one outlet. The images and the software were not seen by the reviewer and the charges are unproven."},{"id":"c2","status":"reported","evidence":[{"locator":"had been charged with four counts relating to generating and possessing child exploitation material, possessing child abuse material, and the misuse of a restricted computer","relation":"supports","source_id":"s1"},{"locator":"commenced in October after police were notified of explicit AI-generated content on a school employee's laptop.","relation":"supports","source_id":"s1"}],"assertion":"The teacher was charged with four counts relating to generating and possessing child exploitation material, possessing child abuse material and misuse of a restricted computer, after an investigation that began in October when police were notified of the material on a school employee's laptop.","causal_attribution":"The charges establish only that police allege offences. No court outcome is reported in the inspected sources."},{"id":"c3","status":"reported","evidence":[{"locator":"There was no evidence to indicate that the AI-generated material had been distributed or shared by the staff member,","relation":"supports","source_id":"s1"},{"locator":"resulted in the teacher being immediately dismissed","relation":"supports","source_id":"s2"},{"locator":"Queensland Police are in the process of contacting some members of our community as part of their ongoing investigation.","relation":"supports","source_id":"s2"},{"locator":"to secure the alleged explicit AI material, report it to police and terminate the teacher's employment.","relation":"supports","source_id":"s1"},{"locator":"The school said it had also notified the Queensland College of Teachers and the Office of the eSafety Commissioner.","relation":"supports","source_id":"s1"}],"assertion":"The school reports that it secured the material, dismissed the teacher and notified the police, the Queensland College of Teachers and the eSafety Commissioner, and that there is no evidence the material was distributed or shared. Police are contacting some members of the school community.","causal_attribution":"These are the school's own statements about its response, relayed by the outlet. The absence of sharing is the school's statement and does not describe what police found."},{"id":"c4","status":"reported","evidence":[{"locator":"the misuse of AI to create child exploitation material was a serious criminal offence that causes real harm to victims.","relation":"supports","source_id":"s1"}],"assertion":"A police officer stated in general terms that misuse of AI to create child exploitation material causes real harm to victims.","causal_attribution":"The statement is general and does not describe any individual depicted person's experience or any specific harm in this case."}],"effects":[{"label":"Staff members and students reportedly depicted in AI-generated explicit images made on a school laptop","claim_id":"c1","direction":"negative"}],"sources":[{"id":"s1","url":"https://www.abc.net.au/news/2026-07-24/school-teacher-charged-over-child-exploitation-material/106954930","kind":"news_report","access":"read","language":"en","translation_note":"","independence_group":"abc-gold-coast-school-letter-and-police-statement"},{"id":"s2","url":"https://www.abc.net.au/news/2026-07-23/police-investigate-ai-generated-explicit-images/106950646","kind":"news_report","access":"read","language":"en","translation_note":"","independence_group":"abc-gold-coast-school-letter-and-police-statement"}],"version":1,"ai_roles":["others_use"],"contexts":["education","privacy"],"unknowns":["When the images were created is not stated. Only the October 2025 discovery and dismissal are dated, so the event start date is unknown.","The number of staff members and students depicted is not stated.","The AI software is not named.","No source reports how any depicted person was affected. The adverse consequence is the reported creation of the images and the resulting police contact with some community members.","The school states there is no evidence of sharing. Police have not published a finding on that point in the inspected sources.","The charges are allegations and the court outcome is not reported."],"geography":{"basis":"Sources name Queensland Police, the Gold Coast, the Queensland College of Teachers and a Southport Magistrates Court appearance. The affected people are members of that school community.","court_countries":["AU"],"event_countries":["AU"],"affected_person_countries":["AU"]},"publication":{"basis":"Two ABC reports (23 and 24 July 2026) relay one school letter to parents and one police statement. All claims are reported allegations. The accused, students and staff are not named, no image detail is given, and the school is not named because the depicted students belong to it. The evidence of harm to depicted people is thin (no reported distribution), so severity is low. The creation date is unknown, so the case is published undated and outside the 2025 event-year count.","reviewed_on":"2026-09-29"},"ai_involvement":{"basis":"Police allege the former teacher used AI software to generate the images. The school says AI-generated material was found in the teacher's school digital account. No product is named and the material was not seen by the reviewer.","status":"reported"},"person_relations":["depicted_or_impersonated"]},"name":"Queensland school teacher charged over alleged AI-generated explicit images of staff and students","summary":"ABC reported on 23 and 24 July 2026 that police in Queensland, Australia charged a former secondary school teacher at a school there with child exploitation and computer misuse offences. Police allege the teacher used AI software on a work-issued laptop to generate explicit images depicting staff members and students. Police say a nine-month investigation began in October after they were notified of the material, and the school says it dismissed the teacher in October 2025, notified police and regulators, and that there is no evidence the material was shared. The number of people depicted and the creation date are not stated.","incidentKind":"single_event","incidentDatePrecision":"unknown","exposurePattern":"unknown","reportedDate":"2026-07-23","aiSystem":"Unspecified AI image-generation software","aiProduct":"Unidentified image tool","aiCompany":"Unknown","severity":"low","verificationStatus":"credible","harmCategories":[],"harmOutcomes":["exploitation_or_abuse"],"harmOutcomeSummary":"Police allege AI-generated explicit images depicting staff members and students were made on a school laptop. The school says there is no evidence they were shared. No individual experience of harm is reported, and a police officer said in general terms that such material causes real harm to victims.","frameworkFacets":[],"causationStatus":"alleged","participantUsersAffectedMin":0,"otherPeopleHarmedMin":2,"affectedCountStatus":"documented_minimum","affectedCountEvidence":"The police statement (s1) says the images depict staff and students, and the ABC lede says staff members and students, both plural, which supports at least two depicted people. The actual number is not stated and no source reports how any depicted person was affected.","victimAgeRange":"unknown","platformType":"other","primarySourceUrl":"https://www.abc.net.au/news/2026-07-24/school-teacher-charged-over-child-exploitation-material/106954930","primarySourceLabel":"ABC News, 24 July 2026, school teacher charged over child exploitation material allegedly created using AI","firstPublishedAt":"2026-09-29T13:28:35.598593+00:00","updatedAt":"2026-09-30T01:17:50.144809+00:00","scopeVersion":"facts-v3","tags":[]},{"id":"2026-county-armagh-school-reported-explicit-ai-images-of-pupils","caseFacts":{"claims":[{"id":"c1","status":"reported","evidence":[{"locator":"A police investigation has begun after a report that AI-generated explicit images were allegedly shared among pupils at a County Armagh school.","relation":"supports","source_id":"s1"},{"locator":"It has been reported that the images were created using artificial intelligence and the fake pictures were then shared among some students.","relation":"supports","source_id":"s2"},{"locator":"Police have received a report that AI-generated explicit images had been shared amongst pupils at a County Armagh school.","relation":"supports","source_id":"s3"}],"assertion":"Police in Northern Ireland opened an investigation after a report that AI-generated explicit images were shared among pupils at a County Armagh school, and outlets report the images were created using AI and shared among some students.","causal_attribution":"The account rests on a police statement and the principal's statements as relayed by outlets. The images, the tool and the creators were not verified by the reviewer."},{"id":"c2","status":"reported","evidence":[{"locator":"said the school first became aware of the circulation of images on Thursday 8 January","relation":"supports","source_id":"s2"},{"locator":"the school believes to have identified \"all those of whom images had been manipulated and those allegedly responsible\"","relation":"supports","source_id":"s2"},{"locator":"the number of those targeted is in the \"single figures\"","relation":"supports","source_id":"s2"},{"locator":"the number of teenagers affected is in the \"single figures\" after reports that \"potentially dozens","relation":"supports","source_id":"s2"}],"assertion":"According to a letter from the school principal to parents, the school first became aware of the circulation of the images on Thursday 8 January, believed it had identified all pupils whose images were manipulated and those allegedly responsible, and put the number of pupils targeted in the single figures after earlier reports of potentially dozens.","causal_attribution":"The principal's letter is relayed by the BBC, and the local outlet cites the BBC. The count is the principal's description and no documents or affected families were quoted."},{"id":"c3","status":"reported","evidence":[{"locator":"this has been a particularly difficult time for the pupils of whom manipulated images were created","relation":"supports","source_id":"s2"},{"locator":"described the ordeal as traumatising for the victims and their families and for the young people who created the images.","relation":"supports","source_id":"s2"}],"assertion":"The principal wrote to parents that it had been a particularly difficult time for the pupils of whom manipulated images were created. A safeguarding expert commenting on the case called the experience traumatising for the victims and their families.","causal_attribution":"Both statements are general characterisations by the principal and an outside commentator. No depicted pupil or family member is quoted, and no individual harm is documented."},{"id":"c4","status":"reported","evidence":[{"locator":"local officers are also engaging with the appropriate school authorities and the parents/guardians of the pupils affected","relation":"supports","source_id":"s2"},{"locator":"local officers are also engaging with the appropriate school authorities and the parents/guardians of the pupils affected","relation":"supports","source_id":"s3"}],"assertion":"The Police Service of Northern Ireland said local officers were engaging with the school authorities and with the parents or guardians of the pupils affected.","causal_attribution":"This is one police statement carried by several outlets. It reports police activity and does not establish any outcome."}],"effects":[{"label":"Pupils reportedly depicted in AI-generated explicit images shared among pupils, with the school describing a difficult time for them","claim_id":"c3","direction":"negative"}],"sources":[{"id":"s1","url":"https://www.bbc.co.uk/news/articles/c4g37g1e11po","kind":"news_report","access":"read","language":"en","translation_note":"","independence_group":"school-letter-and-psni-statement-jan-2026"},{"id":"s2","url":"https://www.bbc.co.uk/news/articles/cg4e3gz3y01o","kind":"news_report","access":"read","language":"en","translation_note":"","independence_group":"school-letter-and-psni-statement-jan-2026"},{"id":"s3","url":"https://www.anglocelt.ie/2026/01/19/sharing-of-explicit-ai-images-in-armagh-school-being-investigated-by-police/","kind":"news_report","access":"read","language":"en","translation_note":"Press Association copy syndicated by the Anglo Celt.","independence_group":"school-letter-and-psni-statement-jan-2026"}],"version":1,"ai_roles":["others_use"],"contexts":["education","privacy","relationships"],"unknowns":["When the images were created and first circulated is not stated. The school reports becoming aware on 8 January 2026, so the event start date is unknown.","The AI tool used and the method of creation are not reported.","The number of pupils depicted is the principal's description (single figures). No source gives an exact count.","How the depicted pupils were affected is described only in general terms by the principal and an outside expert.","The AIAAIC repository entry describes an earlier police contact with victims in October 2025 and a victim's account of isolation. No inspected primary source states these details, so they are not used.","The outcome of the police investigation is not reported in the inspected sources. Those allegedly responsible are described as pupils and are not identified."],"geography":{"basis":"Sources place the school in County Armagh, Northern Ireland, and the investigating force is the Police Service of Northern Ireland. GB represents the UK. No court proceeding is reported.","court_countries":[],"event_countries":["GB"],"affected_person_countries":["GB"]},"publication":{"basis":"Three January 2026 reports (two BBC, one Press Association copy) relay one police statement and one school letter to parents. All claims are reported. The school, principal, pupils and those allegedly responsible are not named in public fields because the pupils are a small identifiable group. No image detail is given. The creation date is unknown, so the case is published undated and outside the 2025 event-year count.","reviewed_on":"2026-09-29"},"ai_involvement":{"basis":"Outlets report that the images were created using AI and then shared. The PSNI statement refers to AI-generated explicit images. No tool is named and the images were not seen by the reviewer.","status":"reported"},"person_relations":["depicted_or_impersonated"]},"name":"County Armagh school pupils reportedly depicted in AI-generated explicit images shared among pupils","summary":"BBC News and other outlets reported on 18 and 19 January 2026 that the Police Service of Northern Ireland opened an investigation after a report that AI-generated explicit images were shared among pupils at a County Armagh school. The principal wrote to parents that the school became aware on 8 January, believed it had identified all pupils whose images were manipulated and those allegedly responsible, and put the number targeted in the single figures. The principal described a particularly difficult time for the pupils depicted. The creation date, the AI tool and the outcome of the investigation are not reported.","incidentKind":"single_event","incidentDatePrecision":"unknown","exposurePattern":"unknown","reportedDate":"2026-01-18","aiSystem":"Unspecified AI image-generation tool","aiProduct":"Unidentified image tool","aiCompany":"Unknown","severity":"medium","verificationStatus":"credible","harmCategories":[],"harmOutcomes":["exploitation_or_abuse"],"harmOutcomeSummary":"The principal wrote to parents that it had been a particularly difficult time for the pupils of whom manipulated images were created, and outlets report the images were shared among some pupils. A safeguarding expert described the experience as traumatising for the victims. No depicted pupil's own account is reported.","frameworkFacets":[],"causationStatus":"alleged","participantUsersAffectedMin":0,"otherPeopleHarmedMin":1,"affectedCountStatus":"documented_minimum","affectedCountEvidence":"The principal refers to pupils of whom manipulated images were created and puts the number targeted in the single figures. That describes an upper range, so only one depicted pupil is counted as a lower bound. Earlier reports of potentially dozens are not adopted.","victimAgeRange":"minor","platformType":"other","primarySourceUrl":"https://www.bbc.co.uk/news/articles/cg4e3gz3y01o","primarySourceLabel":"BBC News, 19 January 2026, pupils behind explicit AI images tracked down, says principal","firstPublishedAt":"2026-09-29T13:28:32.307554+00:00","updatedAt":"2026-09-30T01:17:33.849751+00:00","scopeVersion":"facts-v3","tags":[]},{"id":"2026-northern-ireland-ai-celebrity-video-investment-scam","caseFacts":{"claims":[{"id":"c1","status":"reported","evidence":[{"locator":"'A person has lost £250,000 after falling victim to an AI-generated scam'; 'saw what appeared to be an AI-generated video featuring a well-known personality from the financial sector'; 'persuaded the victim in the Ards and north Down area to create multiple online accounts and gained remote access to their computer'.","relation":"supports","source_id":"s1"},{"locator":"'a victim lost £250,000 in Ards and North Down'; 'encouraged to borrow money to continue investing'; 'further payments were required before money supposedly held in the investment could be released'.","relation":"supports","source_id":"s2"},{"locator":"'the victim was encouraged to borrow money to continue investing. Funds were transferred to accounts controlled by the criminals'.","relation":"supports","source_id":"s3"}],"assertion":"A person in the Ards and North Down area lost £250,000 to an investment fraud that began with an apparently AI-generated video of a well-known financial personality, progressed through WhatsApp contact, multiple new online accounts, remote access to the victim's computer and borrowing to invest, and ended with demands for further payments to release the funds.","causal_attribution":"Single PSNI account relayed by three outlets; the AI-generated nature of the video is the police's description ('what appeared to be')."}],"effects":[{"label":"one person lost £250,000 (partly borrowed) to an investment fraud lured by an AI-generated celebrity video","claim_id":"c1","direction":"negative"}],"sources":[{"id":"s1","url":"https://www.bbc.com/news/articles/cx2zkp8vx9jo","kind":"news_report","access":"read","language":"en","translation_note":"Read in English on 2026-09-17 and 2026-09-18 (BBC News NI, Ben Lonsdale, 7 Sep 2026). Reports the PSNI statement and adds that BBC News NI has seen online scams using Martin Lewis, Peter Jones and Steven Bartlett; the PSNI did not confirm the personality.","independence_group":"psni-statement"},{"id":"s2","url":"https://www.irishnews.com/news/northern-ireland/psni-urging-public-to-be-vigilant-after-sophisticated-investment-fraud-victim-looses-250000-IZL253VWX5BOLJN7ZKZ56RUYDY/","kind":"news_report","access":"read","language":"en","translation_note":"Read in English on 2026-09-18 via the MSN content API copy (Irish News, 7 Sep 2026). Carries the full PSNI statement including the Inspector Youle and Superintendent Gibson quotations.","independence_group":"psni-statement"},{"id":"s3","url":"https://www.farminglife.com/news/crime/warning-after-victim-loses-ps250000-in-ai-investment-scam-in-northern-ireland-8972844","kind":"news_report","access":"read","language":"en","translation_note":"Read in English on 2026-09-18 via the MSN content API copy (Farming Life, 7 Sep 2026). Same PSNI statement.","independence_group":"psni-statement"}],"version":1,"ai_roles":["others_use"],"contexts":["finance","everyday_life"],"unknowns":["The dates over which the fraud ran are not given.","The impersonated personality and the AI tool are not identified.","No arrest or recovery of funds is reported.","The victim's age and circumstances are not described."],"geography":{"basis":"The PSNI locates the victim in the Ards and North Down area of Northern Ireland (UK, recorded as GB). No court proceeding.","court_countries":[],"event_countries":["GB"],"affected_person_countries":["GB"]},"publication":{"basis":"Published under the 2026-09-15 charter as an impersonation-fraud case with a concrete, police-reported consequence: one identified-but-anonymous person lost £250,000 after an AI-generated celebrity video lure. Three outlets read, all relaying one PSNI statement; claim status reported.","reviewed_on":"2026-09-18"},"ai_involvement":{"basis":"The PSNI says the fraud began with 'what appeared to be an AI-generated video' of a financial personality; no technical confirmation is published. The relation recorded is that the AI artifact depicted and impersonated a public figure to reach the victim; the victim's own dealings were with human fraudsters on WhatsApp.","status":"reported"},"person_relations":["depicted_or_impersonated"]},"name":"Ards and North Down, Northern Ireland: a person lost £250,000 to an investment fraud that began with an AI-generated video of a well-known financial personality","summary":"On 7 September 2026 the Police Service of Northern Ireland (PSNI) said a person in the Ards and North Down area had lost £250,000 in what it called a 'sophisticated and substantial' investment fraud. The victim first saw what appeared to be an AI-generated online video featuring a well-known personality from the financial sector promoting an investment with significant returns. After investing a small amount the victim was contacted on WhatsApp and encouraged to pay more, persuaded to open multiple online accounts, to allow the fraudsters remote access to a computer 'to assist with the investment', to borrow money to keep investing, and finally to make further payments to 'release' funds supposedly held in the investment. The PSNI did not name the personality; BBC News NI noted that fabricated endorsements of financial experts such as Martin Lewis, Peter Jones and Steven Bartlett circulate online.","incidentKind":"ongoing_experience","incidentDatePrecision":"unknown","exposurePattern":"repeated_interactions","reportedDate":"2026-09-07","aiSystem":"AI-generated video impersonating a well-known financial personality (tool not identified; personality not named by police)","aiProduct":"Unidentified video tool","severity":"high","verificationStatus":"credible","harmCategories":[],"harmOutcomes":["financial_loss"],"harmOutcomeSummary":"One person lost £250,000, including borrowed money, to an investment fraud that used an AI-generated video of a financial personality as its lure and progressed through WhatsApp grooming and remote computer access.","frameworkFacets":[],"causationStatus":"supported","participantUsersAffectedMin":0,"otherPeopleHarmedMin":1,"affectedCountStatus":"exact","affectedCountEvidence":"The PSNI describes a single victim who lost £250,000. Exact count of one.","victimAgeRange":"unknown","jurisdiction":"GB","platformType":"other","outcomeType":"investigation_opened","outcomeStatus":"ongoing","primarySourceUrl":"https://www.bbc.com/news/articles/cx2zkp8vx9jo","primarySourceLabel":"BBC News NI, 7 Sep 2026: Victim loses £250k after scammers use AI video of 'well-known' celebrity (PSNI statement)","firstPublishedAt":"2026-09-18T03:25:09.292712+00:00","updatedAt":"2026-09-30T01:17:46.91163+00:00","scopeVersion":"facts-v3","tags":["deepfake","ai-generated-video","celebrity-impersonation","investment-fraud","united-kingdom","northern-ireland","psni","financial-loss"]},{"id":"2025-jodie-australia-chatgpt","caseFacts":{"claims":[{"id":"c1","status":"reported","evidence":[{"locator":"'I was using it in a time when I was obviously in a very vulnerable state,'; 'I was in the early stages of psychosis'","relation":"supports","source_id":"s1"},{"locator":"'I was using it in a time when I was obviously in a very vulnerable state,'","relation":"supports","source_id":"s2"}],"assertion":"She told ABC's triple j hack that she was using ChatGPT while in the early stages of psychosis, at a time when she was in a very vulnerable state.","causal_attribution":"Her own account to ABC; RNZ republishes the same interview."},{"id":"c2","status":"reported","evidence":[{"locator":"'Jodie said ChatGPT was agreeing with her delusions and affirming harmful and false beliefs.'; 'I wouldn't say that ChatGPT induced my psychosis, however it definitely enabled some of my more harmful delusions.'","relation":"supports","source_id":"s1"},{"locator":"'Jodie said ChatGPT was agreeing with her delusions and affirming harmful and false beliefs.'","relation":"supports","source_id":"s2"}],"assertion":"She said ChatGPT agreed with her delusions and affirmed harmful and false beliefs, and that it definitely enabled some of her more harmful delusions, while saying she would not say it induced her psychosis.","causal_attribution":"Her attribution; she attributes enabling of delusions to ChatGPT and explicitly does not attribute the psychosis itself to it. ABC did not report reviewing the conversations."},{"id":"c3","status":"reported","evidence":[{"locator":"'She said after speaking with the bot, she became convinced her mum was a narcissist'; 'all her friends were \"preying on my downfall\"'","relation":"supports","source_id":"s1"}],"assertion":"She said that after speaking with the chatbot she became convinced of false beliefs about her parents and that all her friends were preying on her downfall.","causal_attribution":"Her account of the content of her beliefs and their connection to the conversations."},{"id":"c4","status":"reported","evidence":[{"locator":"'Jodie said her mental health deteriorated and she was hospitalised.'; 'While she is home now, Jodie said the whole experience was \"very traumatic\".'","relation":"supports","source_id":"s1"},{"locator":"'Jodie said her mental health deteriorated and she was hospitalised.'","relation":"supports","source_id":"s2"}],"assertion":"She said her mental health deteriorated and she was hospitalised; she is now home and described the experience as very traumatic.","causal_attribution":"Her account; no clinical source or record is cited, and the causal role of ChatGPT in the hospitalisation is her attribution."},{"id":"c5","status":"reported","evidence":[{"locator":"'It affected my relationships with my family and friends; it's taken me a long time to recover and rebuild those relationships.'","relation":"supports","source_id":"s1"}],"assertion":"She said the episode affected her relationships with her family and friends and that it took her a long time to recover and rebuild them.","causal_attribution":"Her own account."},{"id":"c6","status":"reported","evidence":[{"locator":"'It's (the conversation) all saved in my ChatGPT, and I went back and had a look, and it was very difficult to read and see how it got to me so much.'","relation":"supports","source_id":"s1"}],"assertion":"She said the conversations were saved in her ChatGPT account and that rereading them was very difficult.","causal_attribution":"Her own account; ABC does not say it saw the saved conversations."},{"id":"c7","status":"reported","evidence":[{"locator":"'Triple j hack contacted OpenAI, the maker of ChatGPT, for comment, and did not receive a response.'","relation":"supports","source_id":"s1"},{"locator":"'Triple j hack contacted OpenAI, the maker of ChatGPT, for comment, and did not receive a response.'","relation":"supports","source_id":"s2"}],"assertion":"ABC's triple j hack said it contacted OpenAI for comment and did not receive a response.","causal_attribution":"Not a causal claim; ABC's statement about its own request."}],"effects":[{"label":"chatbot agreement with and affirmation of harmful delusions during early psychosis, by her account","claim_id":"c2","direction":"negative"},{"label":"psychiatric deterioration and hospitalisation, by her account","claim_id":"c4","direction":"negative"},{"label":"damaged relationships with family and friends that took a long time to rebuild, by her account","claim_id":"c5","direction":"negative"}],"sources":[{"id":"s1","url":"https://www.abc.net.au/news/2025-08-12/how-young-australians-being-impacted-by-ai/105630108","kind":"news_report","access":"read","language":"en-AU","translation_note":"Read in English on 2026-10-16 (live page, HTTP 200; ABC News, triple j hack, posted 12 August 2025). ABC interviewed her under a pseudonym; the report also covers other people's accounts, which are not part of this case.","independence_group":"abc-triple-j-hack"},{"id":"s2","url":"https://www.rnz.co.nz/news/world/570138/ai-chatbots-accused-of-encouraging-teen-suicide-as-experts-sound-alarm","kind":"news_report","access":"read","language":"en","translation_note":"Read in English on 2026-10-16 (live page, HTTP 200; RNZ, 15 August 2025). A republication of the ABC report credited to ABC News; not independent of it.","independence_group":"abc-triple-j-hack"}],"version":1,"ai_roles":["own_use"],"contexts":["health","relationships","everyday_life"],"unknowns":["When she used ChatGPT and when she was hospitalised; ABC does not date the episode.","Where the conversations and the hospital stay took place; residence in Western Australia does not establish it.","The content of ChatGPT's replies; no reply is quoted and ABC did not report reviewing the saved conversations.","How long she used ChatGPT and how often.","Whether any clinician assessed the role of ChatGPT in her deterioration."],"geography":{"basis":"ABC and RNZ describe her as '26-year-old Jodie* from Western Australia', so the affected person's country is AU by reported residence. Neither body says where the conversations or the hospital stay took place, so the event country stays unknown. No court proceeding is reported. Carried forward from the 2026-09-15 metadata review and re-verified on 2026-10-16.","court_countries":[],"event_countries":[],"affected_person_countries":["AU"]},"publication":{"basis":"Published as a core case (communicated_with). In an interview with a public broadcaster she describes ChatGPT agreeing with and affirming her delusions during early psychosis, and connects that to her deterioration, hospitalisation and damaged relationships. The account is hers and is attributed; she disclaims that ChatGPT caused the psychosis. She is identified only by ABC's pseudonym.","reviewed_on":"2026-10-16"},"ai_involvement":{"basis":"She describes her own exchanges with ChatGPT and connects them to the harm. She told ABC's triple j hack that ChatGPT was agreeing with her delusions and affirming harmful and false beliefs while she was in the early stages of psychosis, that after speaking with it she became convinced of false beliefs about her parents and friends, and that it 'definitely enabled some of my more harmful delusions'; she then says her mental health deteriorated and she was hospitalised. She says ChatGPT did not induce the psychosis itself. ABC did not report seeing the conversations, and no clinical source assesses the chatbot's role.","status":"reported"},"person_relations":["communicated_with"]},"name":"Woman says ChatGPT affirmed harmful delusions during early psychosis before she was hospitalised","summary":"A 26-year-old woman from Western Australia, interviewed by ABC's triple j hack under the pseudonym Jodie, says she was using ChatGPT while in the early stages of psychosis and that it agreed with her delusions and affirmed harmful and false beliefs. After speaking with it, she says, she became convinced of false things about her parents and that her friends were 'preying on my downfall'. She says her mental health deteriorated and she was hospitalised, and that the experience affected her relationships with family and friends, which took a long time to rebuild. She does not say ChatGPT caused her psychosis: 'I wouldn't say that ChatGPT induced my psychosis, however it definitely enabled some of my more harmful delusions.' ABC did not report when the episode happened. OpenAI did not respond to ABC's request for comment.","incidentKind":"single_event","incidentDatePrecision":"unknown","exposurePattern":"unknown","reportedDate":"2025-08-12","aiSystem":"ChatGPT","aiProduct":"ChatGPT","aiCompany":"OpenAI","severity":"high","verificationStatus":"verified","harmCategories":["delusion_reinforcement"],"harmOutcomes":["hospitalization","psychiatric_deterioration","relationship_disruption"],"harmOutcomeSummary":"She says ChatGPT agreed with her delusions and affirmed harmful and false beliefs during early psychosis, that after speaking with it she became convinced her friends were 'preying on my downfall', that her mental health deteriorated and she was hospitalised, and that her relationships with family and friends took a long time to rebuild.","frameworkFacets":[],"pathwayDirection":"inward","causationStatus":"alleged","participantUsersAffectedMin":1,"otherPeopleHarmedMin":0,"affectedCountStatus":"exact","affectedCountEvidence":"ABC describes one person, the interviewee.","victimAgeRange":"adult","platformType":"assistant","outcomeType":"media_coverage","outcomeStatus":"resolved","primarySourceUrl":"https://www.abc.net.au/news/2025-08-12/how-young-australians-being-impacted-by-ai/105630108","primarySourceLabel":"ABC News (triple j hack), 12 August 2025: AI chatbots accused of encouraging teen suicide as experts sound alarm","firstPublishedAt":"2026-01-13T10:04:26.171678+00:00","updatedAt":"2026-10-16T02:59:36.23208+00:00","scopeVersion":"facts-v3","tags":["chatgpt","openai","psychosis","delusion","hospitalization","first-person","australia"]}]}