10 Apr 2026South AfricaChatGPT (reported)
South Africa's Department of Communications and Digital Technologies placed two officials on precautionary suspension with immediate effect, a step the director-general confirmed on 30 April 2026, pending an investigation into the Draft National Artificial Intelligence Policy, which had been gazetted for public comment on 10 April and was withdrawn after News24 and others found fictitious academic sources in its reference list. The department said the irresponsible use of AI tools had compromised the integrity of the document. Briefing Parliament's portfolio committee on 26 May, Minister Solly Malatsi said the two officials were suspended for failing to declare that they used AI, that it had been said ChatGPT was used in editing the document, and that there had been massive oversight and non-disclosure around the use of AI in the policy and its references. The director-general told the committee that the officials had used AI to translate a document from Chile from Spanish into English and that the references came from that document. A new expert panel is guiding a redraft. The officials are not named and the outcome of the investigation is not reported in the sources read.
Core concern Low reported severity Internal Action
AI involvement supported · Causal attribution supported · 6 sources, 4 underlying accounts · Added 17/10/2026
1 Jul 2026 to 14 Oct 2026United KingdomAI Andi
Newbury Street Practice, a GP surgery in Wantage, Oxfordshire, began routing every incoming call through the AI Andi triage system in July 2026. In a Local Democracy Reporting Service article published by BBC News on 14 October 2026, one patient says she waited in the phone queue for 20 minutes before the AI system decided it could not help her, and described the system as a wall that puts patients off contacting the practice. Another patient says she was left frustrated when the system took her call to be about pregnancy. A practice spokesperson said most feedback had been positive (79% in the first month), that some patients less confident with technology had found the change harder, and that patients are now offered the option of speaking to a human receptionist. The local MP said residents had contacted him about the system.
Core concern Low reported severity
AI involvement reported · Causal attribution alleged · 2 sources, 1 underlying account · Added 17/10/2026
29 Jan 2026NetherlandsUnidentified AI writing tool
A law student at Leiden University used a generative AI tool to spell-check his bachelor thesis and to check and order its reference list. Seven of the thesis's footnote references (one of them appearing twice, out of 36 footnotes with references) were wrong in the final version; according to his explanation, the tool had distorted some of them, for example with wrong years or author initials, and invented others. On 29 January 2026 the faculty's exam board declared the thesis invalid and excluded him until 28 January 2027 from the thesis track and every exam in the faculty's programmes, reasoning that the fictitious references indicated unauthorised use of generative AI and therefore fraud. The university's appeals board for examinations found in March 2026 that he had plausibly shown he used generative AI to prepare his reference list, which it called a permitted use, but held him responsible for the non-existent sources; the exam board then re-imposed the invalidation with an exclusion until 29 July 2026, and the appeals board upheld that in July. On 14 October 2026 the Administrative Jurisdiction Division of the Council of State (Raad van State) found his explanation plausible, held that no more than carelessness had been shown, and annulled the sanction; the thesis must now be graded, and the examiner may take the incorrect references into account in the grade.
Core concern Medium reported severity
AI involvement supported · Causal attribution supported · 2 sources, 1 underlying account · Added 17/10/2026
31 Mar 2026 to 11 Sept 2026United StatesIrys (reported)
A Los Angeles County Superior Court judge ordered a senior associate at Musick, Peeler & Garrett, the outside firm defending State Farm in a homeowner's lawsuit over a fire-damaged home, to pay a $999.99 sanction after the associate's motions in limine cited cases, quotations and holdings that did not exist, Law360 reported on 14 September 2026. In a declaration reproduced by Reason's Volokh Conspiracy, the associate said she used an AI program called Irys to prepare the motions, did not check some of the citations, and wrongly believed the program was cite-checked through the firm's Westlaw subscription; she confirmed seven non-existent case citations across eight motions. The plaintiff's lawyers found the citations and raised them at a status conference on 7 August 2026. The firm acknowledged citations that appeared to be AI hallucinations, and its lead trial counsel apologised to the court and the plaintiff. According to Law360, the judge had considered up to $5,000, and the amount set is one cent below the level at which California law requires a court to notify the State Bar.
Core concern Low reported severity Regulatory Action
AI involvement supported · Causal attribution supported · 4 sources · Added 17/10/2026
Jul 2026Event location unknownUnidentified image tool
A 23-year-old woman told the news site Kaktus.media that in July 2026 she learned that acquaintances had received a nude image of her made with AI from her photograph, and that the sender was a classmate. She says he first told her an app had generated and sent the image by itself, then admitted that he had undressed her Instagram photo through AI himself. In a written statement to the outlet the accused says he used a Telegram bot offering AI photo processing in May or June 2022 and was shocked by the result, and that the images were sent from his Telegram account on the night of 11 to 12 July 2026 by people who had hacked it. The woman says she suffered severe psychological stress and harm to her reputation. She filed a complaint with a district police department on 21 July; after she complained to the prosecutor's office the matter moved to another district department. That department told Kaktus.media that, because the review period expired before expert results arrived, it had refused to open a criminal case and had ordered a new examination. Her lawyer says four motions filed on 5 August have gone unanswered.
Core concern Medium reported severity Investigation Opened
AI involvement reported · Causal attribution supported · 1 source · Added 17/10/2026
14 Oct 2026United StatesUnidentified image tool
According to the Hillsborough County Sheriff's Office, as reported by CBS News Miami, WFLA and WTSP, deputies responded to a high school in Riverview, Florida, on 14 October 2026 after administrators received an email containing a screenshot of an Instagram post threatening a shooting at the school. The post appeared to come from a male student. Detectives interviewed him and determined that the post did not come from him. The sheriff's office's communications chief said detectives very quickly realised the post was AI-altered and linked it to his ex-girlfriend, a teenage former student. Police say she admitted sending the email knowing that the image had been altered using AI. She was arrested and charged with making a written or electronic threat to conduct a mass shooting or act of terrorism, unlawful use of a two-way communications device and disrupting a school campus or function. The tool used and how detectives established the alteration are not reported.
Core concern Low reported severity Criminal Charges
AI involvement reported · Causal attribution alleged · 3 sources, 1 underlying account · Added 17/10/2026
Jan 2026ChinaUnidentified AI search tool
Zhejiang Provincial Tongde Hospital in Hangzhou told Red Star News (relayed by China News Service on 26 January 2026) that a patient with a cold sore had recently used AI online to look up treatment, bought valaciclovir and took it as advised. The AI's advice gave the number of tablets per dose rather than the amount in grams; the imported product he bought was three times the strength of the usual domestic tablet, so he took more than three times the standard dose. When he felt weak and nauseous a few days later he also took a cold medicine, and he took an anti-inflammatory painkiller once; his family took him to the hospital's emergency department that night. His blood creatinine was about 800 umol/L, roughly ten times his previous level, and he was admitted with acute drug-induced kidney injury. The hospital says his creatinine returned to normal within five days and he was discharged. The AI tool and its exact wording are not identified.
Core concern Medium reported severity Media Coverage
AI involvement reported · Causal attribution alleged · 2 sources, 1 underlying account · Added 16/10/2026
Oct 2026Event location unknownClaude Code (reported)
A GitHub issue filed on 15 October 2026 against Claude Code (version 2.1.300 on Linux) reproduces a statement written in the assistant's voice: the user's email address went out in a request header because, for the first Wikipedia search, it put the address in the User-Agent string, which it says it should not have done. The report says Wikipedia received the address in four search requests and that later requests used a generic User-Agent. Version 2.1.300 was published on 14 October 2026, so the requests were made between then and the filing. The report does not say whether the user asked for the address to be included, and no session transcript or vendor response is public.
Core concern Low reported severity
AI involvement reported · Causal attribution alleged · 3 sources · Added 16/10/2026
Oct 2026BrazilUnidentified content-generation tool
g1 Santarém reported on 15 October 2026 that the Pará Civil Police had located a 70-year-old woman at Santarém airport, in western Pará, after she was reported missing. According to the Civil Police, she had left home intending to travel to meet a person she had been talking to online, and she was being deceived through content produced with artificial intelligence that promised her a job, financial advantages and a romantic relationship. A police delegate told g1 that the woman believed the profile's promises that she could work, earn money and even marry the person, and that she did not report being asked for money. g1 does not report how the police established that the content was made with AI or which platform or tool was involved. The police warned that older people travelling alone to meet online contacts can be exposed to other crimes or go missing without their families knowing where they are.
Core concern Low reported severity Media Coverage
AI involvement reported · Causal attribution alleged · 1 source · Added 16/10/2026
12 Aug 2026ChinaUnidentified chatbot
On the evening of 12 August 2026 rescuers were called to Tonggongjian, a peak of about 1,558 metres in the undeveloped mountains of Lin'an district, Hangzhou, where a man who had climbed alone had fallen and was stranded at about 1,300 metres. A volunteer rescue team with police and firefighters, and in the fire service's account its firefighters, found him after more than an hour, with abrasions, and escorted him down. Dushi Kuaibao's report of the rescue says the man had argued with his wife, asked an AI app on his phone where to go, and that the app asked about his state of mind, personality, height and fitness before recommending Tonggongjian; the fire service's account says it suggested he go there to watch a meteor shower. The rescue team said he rarely hiked, wore flat skate shoes and carried only a folding stool, and that the recommendation gave no advice on equipment. The app is not named and the exchange is not published.
Core concern Low reported severity Media Coverage
AI involvement reported · Causal attribution alleged · 3 sources, 2 underlying accounts · Added 16/10/2026
2026SpainUnidentified AI drafting tool
The Criminal Chamber of the High Court of Justice of the Canary Islands (TSJC) fined a lawyer 420 euros for breaching the duty of truthfulness and good faith, after an appeal the lawyer filed contained 48 citations of Supreme Court judgments and a report of the General Council of the Judiciary that did not exist. According to the court's order, as described in its press release of 13 February 2026, the lawyer used a general-purpose AI tool to draft the appeal and did not check a single citation against the free public case-law database; the court's documentalist verified that the references were false. The court referred the matter to the lawyer's bar association for possible disciplinary action and set the fine at about half the annual cost of a legal-specific AI tool, taking into account that the lawyer admitted the facts and expressed remorse. The court had first flagged the citations in its judgment dismissing the appeal, reported on 5 January 2026, which opened a separate proceeding. The lawyer told the court of comments and mockery from colleagues after the case became public.
Core concern Low reported severity Regulatory Action
AI involvement supported · Causal attribution established · 3 sources, 2 underlying accounts · Added 16/10/2026
9 Oct 2026 to 13 Oct 2026BrazilUnidentified video tool
In October 2026 a 49-year-old builder travelled more than 2,000 km by bus from Imperatriz, in Maranhão, to Belo Horizonte to meet a woman he knew from a social-media account with about 300,000 followers. In the account's videos a woman with a realistic voice and appearance said she was looking for a man to live with her on a farm. He told G1 he began interacting with the profile after its videos appeared in his feeds, that it kept private conversations with him and offered to pay his fare, and that he travelled without an address. In Belo Horizonte he showed her photos to people and found that nobody knew her; he said he felt deceived. The city's social assistance service gave him food, a bath and a ticket home. A day later the account's administrator posted a video in which the persona says it is made of artificial intelligence, that it never conversed with anyone on the profile and that the aim was engagement for sales; the creator told the press he set up the account in 2025 to attract buyers. Who wrote the private messages the man describes is not established. Minas Gerais civil police said they had no report of the case and could not say a crime had occurred.
Core concern Low reported severity Media Coverage
AI involvement supported · Causal attribution alleged · 3 sources, 2 underlying accounts · Added 16/10/2026
5 Feb 2026 to 9 Oct 2026United StatesvLex, Paxton AI and NotebookLM (reported)
In a federal trademark suit in the U.S. District Court for the Southern District of New York, the court on 5 February 2026 struck a company defendant's motion papers and entered default judgment against it as a sanction on its lawyer, after his filings repeatedly cited nonexistent cases and misattributed quotations. The lawyer said he had found and cite-checked cases with AI tools (vLex and Paxton AI) and used NotebookLM to review a later submission; the court rejected his explanations and found he acted in bad faith. On 9 October 2026 the court described the citations as produced by generative AI and ordered the lawyer to pay $38,877.59 in fees and costs to the firm of a co-defendant's former lawyer, who had reported the false citations to the court.
Core + contextual relations Medium reported severity
AI involvement supported · Causal attribution established · 2 sources, 1 underlying account · Added 16/10/2026
Jun 2026 to Sept 2026GermanyUnidentified image tool (suspected)
The Gauting police inspection in Bavaria, Germany, reported on 2 October 2026 that a man had filed a complaint after realising he had been the victim of investment fraud. According to the police report, he said he had clicked on an advertising banner for an investment on the website of a large German news magazine, and the police say he was presumably lured by the combination of an AI-generated image of a well-known television presenter with a supposedly lucrative investment strategy. After he entered his email address and phone number on the trading platform, four women callers contacted him from 4 June 2026 until September and led him to make four payments totalling more than 25,000 euros to accounts in Bulgaria, Malta and Lithuania. His bank could not reverse the transfers because he had authorised them himself. The police do not say how they concluded that the image was AI-generated, and the presenter is not named in the police report.
Core concern Medium reported severity Investigation Opened
AI involvement suspected · Causal attribution alleged · 2 sources, 1 underlying account · Added 16/10/2026
13 Oct 2026BangladeshUnidentified video tool
On 13 October 2026 two videos depicting two Bangladeshi actresses in sexual content were posted on the official Instagram account of the private television channel Maasranga Television. The fact-checking organisation Rumor Scanner reported, as relayed by Prothom Alo and Dhaka Files, that the videos were not genuine and had been made by placing the actresses' faces onto other videos with face-swap technology; Dhaka Files adds that the source videos came from adult websites and describes the technology as AI. The channel said its account had been hacked and that fake images and videos made by misusing AI had been published; it announced that it had recovered the account with the help of BTRC, the CID and Banani police station, removed the videos, expressed regret and said it would take legal action. Dhaka Files reports the videos were viewed roughly 400,000 times before deletion and records doubts about the hack claim, because routine posts continued from the account during the period. On 15 October one of the actresses wrote on her Facebook page that creating and spreading fabricated content in someone's name with AI is a crime. Who made or posted the videos is not established in the sources.
Core concern Medium reported severity Media Coverage
AI involvement reported · Causal attribution supported · 2 sources · Added 16/10/2026
Apr 2026 to Jun 2026HondurasUnidentified image tool
Honduras's National Police announced on 14 October 2026 that its anti-kidnapping unit (UNAS) of the Police Investigations Directorate (DPI) had arrested a woman, under a court warrant, on suspicion of simulating a non-existent offence related to kidnapping and of using false documents. According to the police statement, the investigation found that she had faked her own kidnapping twice, in episodes reported on 21 April and 1 June 2026, to obtain money from her partner to pay personal debts, and that she received an initial deposit of 2,000 lempiras after making the partner believe she was being held and that her captors were demanding money. Police say the agents determined that she presumably used artificial-intelligence tools to generate false images of injuries and bruises, which were sent to the partner to make the partner believe she had been beaten by the supposed kidnappers, together with apparently forged hospital documents. According to the investigation file, she said after being located that she planned both episodes and used AI to make the images. She was brought before the issuing court; no court finding is reported.
Core concern Low reported severity Criminal Charges
AI involvement reported · Causal attribution alleged · 1 source · Added 16/10/2026
2026Event location unknownUnidentified customer-service chatbot
Xinmin Evening News reported on 13 October 2026, in a piece on consumer complaints over the October 2026 National Day holiday, a shopper who paid 12,099 yuan for a branded mobile phone on an unnamed platform; the merchant did not ship it at the agreed time. According to the report, he contacted customer service many times but got only the AI's automatic replies, the same few sentences over and over, and could never get transferred to a human. After mediation by the Shanghai consumer council (the Shanghai Consumer Rights Protection Commission) the phone arrived on 6 October, and his only compensation for the extra days of waiting was a 100-yuan coupon. The council's holiday figures, relayed by Laodong Bao, list looping AI customer-service replies and difficulty reaching a human among the concentrated problems. The platform, merchant and chat record are not reported.
Core concern Low reported severity Media Coverage
AI involvement reported · Causal attribution alleged · 2 sources, 1 underlying account · Added 15/10/2026
2026UkraineUnidentified voice-changing tool
Ukrainian prosecutors, the SBU security service and the National Police announced on 14 October 2026 that they had shut down a fraudulent call centre in the Lviv region of Ukraine. According to the Lviv regional prosecutor's office and the Office of the Prosecutor General, the group created fake women's accounts on dating services and messengers and, to make the conversations more convincing, used software with a neural network that changed a male voice to a female one in real time. Victims were persuaded to download a file disguised as a photograph that contained remote-access malware and to share their screens; the group then took control of their banking apps and moved the money to accounts it controlled or into cryptocurrency. Prosecutors say that, on preliminary data, at least 50 Ukrainian citizens and 30 foreigners, including people in Poland, Germany and Georgia, were victims. Three men were notified of suspicion, including under a new criminal-code article on fraudulent call centres in force since September 2026, and were remanded in custody with the option of bail. The statements do not say how many victims heard the altered voice or how much each lost.
Core concern Medium reported severity Criminal Charges
AI involvement reported · Causal attribution alleged · 3 sources, 1 underlying account · Added 15/10/2026
2026Event location unknownChatGPT (reported)
VnExpress reported on 15 October 2026 that early in the year, just after receiving a Tet bonus, an investor asked ChatGPT which three stocks to buy now with 200 million dong and after several exchanges about needs and risk appetite received three stocks in the banking, information-technology and securities groups. ChatGPT suggested splitting the purchases, waiting for dips and setting take-profit and stop-loss levels, and each answer said it was not an investment offer with committed profit, which VnExpress says the investor disregarded, buying all three at once. All three were in profit more than a month later, and the investor then treated ChatGPT as a professional broker and kept asking it for tips. In recent months the recommended blue-chip stocks fell continuously; the portfolio is in the red, with one stock down 48%, and the investor told VnExpress that for weeks they had stopped checking it. The article presents the account alongside warnings from securities-company staff about relying on general-purpose chatbots for trading decisions.
Core concern Low reported severity Media Coverage
AI involvement reported · Causal attribution alleged · 2 sources, 1 underlying account · Added 15/10/2026
17 Jul 2026United StatesUnidentified voice-cloning tool
The Boone County Sheriff's Office in Missouri said that on Friday 17 July 2026 a resident received a call whose caller ID showed their daughter's number. A voice that sounded like the daughter said she had been kidnapped at gunpoint, and another voice then demanded ransom. The office said it later determined that the voice was AI-generated, without saying how. Deputies found the daughter safe in Jefferson City and unaware of the call. The office called it the first AI voice-impersonation call it had encountered. No suspect had been identified by 19 July, and no source reports whether any money was paid.
Core concern Low reported severity Investigation Opened
AI involvement reported · Causal attribution supported · 3 sources, 1 underlying account · Added 15/10/2026