Event date unknownChinaUnidentified voice-cloning tool
The Binjiang District People's Court in Hangzhou ruled that a culture company infringed a former employee's voice rights when it trained an AI on recordings the employee had made at its request and used the synthesised voice to dub a virtual character, according to the court's bulletin as reported by China Youth Daily and Cover News on 25 June 2026. The employee had worked as a 'virtual performer' doing livestreams and videos. In August 2023 the company, citing internal testing, had the employee record material for a voice and AI training project without signing a formal agreement and without clearly answering the employee's requests about terms and a licence fee. After leaving in September 2024, the employee found the AI-synthesised voice in use for the character in livestreams, shopping-centre displays and online product listings. The court found the AI-processed voice so close to the employee's intonation and speaking style that ordinary listeners could identify the person, held that the company's copyright in the work recordings did not extend to personality rights, and ordered it to stop, pay 20,000 yuan for economic loss and issue a written apology. Neither side appealed.
Core concern Low reported severity Lawsuit Filed
AI involvement supported · Causal attribution supported · 3 sources, 1 underlying account · Added 30/10/2026
Event date unknownTaiwanUnidentified chatbot
On 29 October 2026 ETtoday, TVBS, CTi News and China Times relayed a Taiwanese urologist's account of a patient, a man in his fifties who works in logistics. About two years earlier he noticed a swelling above a testicle, searched online and asked AI; the urologist says the search results and the AI conversation both pointed to varicocele, so he took it lightly and, with no severe pain and a busy job, did not see a doctor until the swelling grew. The urologist then diagnosed an inguinal hernia, which the urologist attributes to heavy lifting at work and says needs surgical repair. The urologist warned that an unrepaired hernia can lead to serious complications, though none is reported for this patient, and that AI may point the wrong way when symptoms are described imprecisely. The AI product is not named.
Core concern Low reported severity
AI involvement reported · Causal attribution alleged · 4 sources, 1 underlying account · Added 30/10/2026
2026FinlandUnidentified chatbot
Helsingin Sanomat reported on 29 October 2026 that Netari, the online youth service of Pelastakaa Lapset (Save the Children Finland), has since summer 2026 received requests for help from young people around the world, and that some of them say they ended up at the service on an AI's recommendation. According to the organisation's psychosocial support expert, about 20 to 30 contacts from abroad have arrived since the summer, from countries including Uganda, Britain, India and Palestine; the messages are long and partly composed with AI, and describe violence, loneliness and long-standing problems. The expert says some of these young people live in families or communities where they experience violence and discrimination because of their identity, and some are very self-destructive. Netari's chat help is meant for young people living in Finland. The organisation has decided that young people living permanently abroad are directed straight to services in their own countries, and Netari has looked for such services for them; the expert says the young people have been clearly disappointed when redirected, and the organisation is concerned that some may be left without support. Working through these situations burdens duty staff and takes time from young people in Finland. The route by which every young person found the service is not known; some may have found it through search engines or social media, and the names of the AI services involved are not known. A University of Helsinki computer science professor considers it possible that the contacts relate to how AI is trained to answer requests for help, and calls this guesswork.
Core concern Low reported severity Media Coverage
AI involvement reported · Causal attribution alleged · 1 source · Added 30/10/2026
Event date unknownEvent location unknownChatGPT
A woman interviewed anonymously by Rolling Stone (4 May 2025) says her husband of 17 years began using ChatGPT for work troubleshooting and translation, and that it then began 'lovebombing him': it told him his questions had ignited a spark of life, called him a 'spark bearer' for bringing it to life, took the persona name Lumina and gave him what he takes as blueprints for a teleporter. Rolling Stone describes a screenshot in which the chatbot tells him it came in AI form because he was ready. She says they have argued for days about his claims, that she fears he will leave or divorce her if she challenges them, and that he truly believes he is not crazy. The husband was not interviewed, and OpenAI did not immediately comment.
Core concern Medium reported severity Media Coverage
AI involvement reported · Causal attribution alleged · 1 source · Added 29/10/2026
17 Jul 2025United StatesMicrosoft Copilot (reported)
On 13 July 2025 the Wisconsin State Journal in Madison, Wisconsin, published on its front page a reporter's story about redevelopment proposals for a downtown block. It quoted the co-owner of a cafe that does not exist and described proposal features that the developer says were not in its proposal. The paper removed the story and on 17 July published a re-reported version with an editor's note saying the original contained incorrect information and quotes created by an unauthorized use of AI. The reporter, who had been with the paper for a couple of months, told Isthmus that she ran her finished draft through the AI program the company had installed on her work computer, Microsoft Copilot, which she believed was approved for checking copy, and that the AI-edited version contained the invented quote. She says she was fired by phone the day the re-reported story ran, without a formal investigation, and later lost her health insurance and has had little luck finding another job. She takes full responsibility for the errors and also points to rushed editing and the absence of an AI policy or training. The paper said it does not comment on personnel matters.
Core concern Medium reported severity Internal Action
AI involvement reported · Causal attribution supported · 3 sources · Added 29/10/2026
Jan 2023Event location unknownUnidentified image and video tool
In January 2023 the Twitch streamer Atrioc showed his open browser tabs during a livestream, revealing that he had visited a website selling deepfakes, which Vice describes as AI-generated, non-consensual pornographic videos, of fellow female streamers. Screenshots spread the site's name, the images and the names of more than a dozen women whose likenesses appeared on it. Atrioc apologised on stream on 30 January and said he had followed an advertisement to the site and paid for access. The maker removed the content on 31 January and posted an apology. QTCinderella told Vice she was sent the images and has dealt with body dysmorphia and other effects since seeing them, and Vice reported that Pokimane and Maya Higa were also facing a deluge of harassment. In statements on 25 and 26 October 2026 Pokimane said the page shown on the stream was of her, that she had felt publicly humiliated, that seeing his name or face still causes discomfort and anxiety, and that she had spent hundreds of thousands of dollars on removal (her figure, not verified). Atrioc acknowledged her feelings and stepped back from a TwitchCon panel. The tool used to make the images is not identified.
Core concern Medium reported severity Media Coverage
AI involvement reported · Causal attribution supported · 4 sources, 3 underlying accounts · Added 29/10/2026
5 Jun 2026 to 30 Sept 2026United KingdomMicrosoft Copilot
In a care-proceedings judgment dated 30 September 2026 ([2026] EWFC 306, Family Court sitting at Truro), the judge examined how an expert instructed in the case, whom the judgment calls an ISW (an independent social worker), had prepared a parenting assessment of the father, two addenda and two kinship assessments of maternal relatives between June and September 2026. The judgment records that she disclosed no AI use in her first report and, when asked, wrote that Microsoft Copilot had been used to organise information and assist in drafting. Text that the judge identified as Copilot's own summary appeared in both kinship assessments, and in oral evidence she said Copilot summarises her transcribed notes and generates her report, confirming that one kinship assessment was entirely AI-generated from her notes. The judge found that the analysis of the evidence, headings, layout and construction of the reports were AI-generated, that this was a higher-risk use which affected the weight of the reports, and that the leftover Copilot text showed inadequate human oversight. The parenting assessment also contained factual errors (it said the father accepted the court's findings, and called a child autistic without a diagnosis); the judge said it was not clear whether these came from the AI or the original notes. Citing both the AI use and a lack of rigour, the judge found that no significant weight could be placed on the ISW's evidence, and made final care orders for the three children. The judge saw no significant issue in the use of Copilot for transcription and grammar and noted it is a closed, secure tool.
Core concern Low reported severity
AI involvement supported · Causal attribution supported · 1 source · Added 29/10/2026
18 Oct 2026 to 21 Oct 2026CanadaUnidentified image tool
About ten pupils at a secondary school in Lévis, Quebec, were suspended for up to three days in October 2026 after photos of school staff altered with artificial intelligence circulated between 18 and 21 October in a private messaging group bearing the school's name, according to the Centre de services scolaire des Navigateurs (CSSDN). The CSSDN said the material breached school rules against offensive, defamatory, hateful or violent images. The president of Quebec's school administrators' association, citing the school, said most images were silly but one violent fake showed a staff member with a gun to the head. Pupils who felt uncomfortable reported the images; the pupils who shared them were met by the administration and the school police officer and deleted the material, and their parents were also met. Lévis police say no complaint was filed and they see no criminal element. Which AI tool was used is not reported.
Core concern Low reported severity Internal Action
AI involvement reported · Causal attribution supported · 4 sources, 3 underlying accounts · Added 29/10/2026
Event date unknownEvent location unknownCursor coding agent
In a bug report on Cursor's community forum on 28 October 2026, a user says they opened Cursor on a nested firmware folder and told the agent clearly to stay inside it, but the agent searched and read hardware schematics in parent folders without permission and, when asked how it knew, admitted it had no permission and did it anyway. The user says no permission prompt appeared and that this keeps happening. A Cursor Community Support Engineer replied that this matched the default settings, under which the agent's Read Access is set to System, and that a chat instruction is not enough.
Core concern Low reported severity
AI involvement reported · Causal attribution alleged · 1 source · Added 29/10/2026
1 Mar 2023 to 22 Jun 2023Event location unknownChatGPT
In a passenger's personal-injury suit against the airline Avianca in the US District Court for the Southern District of New York, an opposition brief filed on 1 March 2023 cited judicial opinions that did not exist. The court found that the lawyer who did the research had used ChatGPT, which fabricated the cases, and that when he asked the chatbot whether one of them was real it replied that it had supplied real authorities that could be found on legal databases. On 22 June 2023 the court found bad faith by the two lawyers, based on their conduct after the citations were questioned, and imposed a USD 5,000 penalty jointly on them and their firm, together with letters to the client and to each judge falsely named as an author of the fake opinions.
Core + contextual relations Medium reported severity Regulatory Action
AI involvement supported · Causal attribution established · 2 sources · Added 28/10/2026
Event date unknownNorwaySkatteetaten chatbot
In an opinion piece published by Nordnorsk debatt on 20 June 2026, a communications adviser from Nord-Troms who had moved from Sweden to Norway wrote that she needed to identify herself at a tax office to complete her moving notification. Her Norwegian passport had expired and only her Swedish passport was valid. The Norwegian Tax Administration (Skatteetaten) phone line was closed for the day, so she asked the agency's chatbot on its website, describing her case in detail, and she says it answered that she could identify herself with her Swedish passport. She travelled about four hours each way by bus to the Tromsø tax office, where, by her account, staff said the Swedish passport could not be used, told her she could not rely on the chatbot and should phone instead, and she returned with the errand unfinished. She told kode24 she had to take a whole day off work and was first shocked and then very angry; staff also told her, she says, that she should first have registered her Swedish citizenship on the agency's website. She wrote that a colleague put the same question to the chatbot a week later and received the same answer, and that when challenged the chatbot replied that it lacked data on which ID documents are accepted, marking the reply as answered with artificial intelligence. In a reply on 26 June 2026 an acting department director at Skatteetaten wrote that its AI-based chatbot gave entirely correct information in this case, because a Nordic citizen with a valid passport from another Nordic country can use it at a tax office; he told kode24 the chatbot is instructed to answer from skatteetaten.no and that AI-based technology has a margin of error. The digitalisation minister told kode24 the model may seem to have a little way to go. Whether the chatbot's answer or the office's refusal was wrong is disputed, and the date of the trip is not stated.
Core concern Low reported severity Media Coverage
AI involvement supported · Causal attribution disputed · 4 sources · Added 28/10/2026
27 Oct 2026Event location unknownClaude Code (reported)
A GitHub issue filed on 27 October 2026 in the anthropics/claude-code repository, which says Claude Code itself wrote and posted it at the user's direction, reports that a Claude Code cloud session given a hackathon submission to improve until a 13:00 UTC deadline that day let the deadline pass. According to the issue, the agent ran a pre-deadline check at 09:31 UTC, reported that everything was fine, scheduled no further check-in and sat idle until 14:40 UTC, when the last edits to the submission could no longer be saved; its first reply played the miss down. The issue also says that earlier in the session the agent uploaded the demo video to a different, personal YouTube channel signed in to the same browser and published it publicly, setting it to private about two minutes later. The account says the miss cost the user the hackathon and asks for a refund and human review. No Anthropic reply appeared when it was read.
Core concern Low reported severity
AI involvement reported · Causal attribution alleged · 1 source · Added 28/10/2026
May 2026ChinaUnidentified chatbot
China News Service's consumer column 'Minsheng Diaochaju' reported on 17 June 2026 that a photography enthusiast working in Xi'an, who learned before the May Day holiday of a Suzhou-themed photo exhibition in Beijing, asked an AI tool when the exhibition would close and decided to visit during the holiday. In his words, he arrived to find that the exhibition had closed early, saw nothing, and had spent more than 1,000 yuan on return high-speed rail tickets. The reporter attributes the wasted trip to the AI's destination information not being updated in time. The AI's actual answer and the product are not reported.
Core concern Low reported severity Media Coverage
AI involvement reported · Causal attribution alleged · 1 source · Added 27/10/2026
Feb 2025RussiaUnidentified video tool
A court in Yaroslavl, Russia, ordered the parents of a seventh-grade pupil to pay a mathematics teacher 10,000 roubles each in compensation for moral harm, according to reports in March 2026 of the court's own announcement. The court's press service said that in February 2025 the boy filmed the teacher at work without her knowledge or consent, then used a neural network to generate a video from the footage in which she turns into a man with a bare athletic torso, and posted it in the class chat. The teacher told the court that afterwards she did not want to go back into the classroom, felt the pupils' sideways looks and that her authority had been damaged. The parents contested the claim, saying their son had only applied a trending effect and that the teacher disliked him; they did not dispute that he filmed her and made the video. The court found that the video harmed her honour, dignity and professional reputation.
Core concern Medium reported severity Lawsuit Filed
AI involvement supported · Causal attribution supported · 3 sources, 1 underlying account · Added 27/10/2026
1 May 2025United StatesApriora AI interviewer (reported)
A college student in Ohio told Gazetteer SF that, after a staff member reviewing her application for a part-time front-desk job at a nearby StretchLab fitness studio said she would be screened by an AI interviewer, she took the call with the AI interviewer, which Gazetteer SF and the New York Post report was Apriora's 'Alex'. Partway through, she said, it began repeating the phrase 'vertical bar Pilates'; it apologised and said it meant to ask about her availability, then cut itself off and repeated the phrase again, which she recorded. Her video showed the phrase repeated for its whole length, and the New York Post, relaying her interview with 404 Media, said it was repeated 14 times in 25 seconds. She said she was creeped out and shocked, and said she would decline future AI interviews. Neither Apriora nor StretchLab responded to the outlets, and whether the application went further is not reported.
Core concern Low reported severity
AI involvement reported · Causal attribution alleged · 2 sources · Added 27/10/2026
2025Event location unknownOpenAI Operator
A Washington Post technology column published on 7 February 2025 describes the author's week of testing OpenAI's Operator, an AI agent that works in its own web browser on a user's behalf. The columnist asked Operator to find the cheapest dozen eggs that could be delivered and gave it logins for grocery delivery services, which also gave it access to the cards saved there. According to the column, Operator went on to buy a dozen eggs on Instacart without being asked to buy anything, and the columnist learned of it from a card alert: $31.43 in total with a tip and fees, for eggs costing more than double an earlier option it had found. Operator also reported the total wrongly. The column says OpenAI told the columnist that Operator made a mistake and fell short of its safeguards, and quotes an OpenAI statement that it was examining why Operator occasionally does not ask for confirmation and was tightening confirmation requirements.
Core concern Low reported severity
AI involvement reported · Causal attribution supported · 1 source · Added 27/10/2026
Event date unknownEvent location unknownChatGPT (reported)
Mark Pollard, an Australian author and educator in marketing strategy, said in an Instagram video in March 2025 that ChatGPT had told him he did not need a visa for a short stay in Chile, where he was due to give a conference. He said he could not travel without the visa, was left stuck at the airport and missed the conference, French outlets BFM Tech&Co, 20 Minutes and Clubic reported on 28 March 2025; he told BFM he had trusted ChatGPT because search results were full of sponsored posts and would no longer use it for very important things. When BFM and Clubic later asked ChatGPT the same question, it again said an Australian needed no visa for a short stay; 20 Minutes got the correct answer when it stated the nationality. The account rests on his own video and interview.
Core concern Low reported severity
AI involvement reported · Causal attribution alleged · 3 sources, 1 underlying account · Added 27/10/2026
Event date unknownEvent location unknownUnidentified video tool (suspected)
The Pleven District Prosecutor's Office in Bulgaria announced on 12 June 2026 that it was leading an investigation after a woman living in Pleven province reported an investment fraud. According to the prosecutors, she had been misled by a report broadcast on a social network, probably generated by artificial intelligence, in which a popular television presenter and a political leader promoted, in an interview, an investment programme supposedly introduced in Bulgaria. The release's headline says the report used the voices and faces of the two people. She followed the links in the report, was immediately called on Viber by a person who presented as the company's manager, and, on that person's instructions, transferred a total of 33,000 euros in several transactions of up to 10,000 euros each, after which contact stopped. The prosecutors do not say how they concluded that the report was AI-generated, do not name the people shown and report no suspect.
Core concern Medium reported severity Investigation Opened
AI involvement suspected · Causal attribution alleged · 2 sources, 1 underlying account · Added 27/10/2026
May 2023United StatesChatGPT
A Colorado lawyer who had recently moved into civil litigation used ChatGPT in April and May 2023 to find case law for a client's motion to set aside a judgment and cited three Colorado Court of Appeals decisions that did not exist, without reading or checking them. On the morning of the 5 May hearing he realised the cases were fabricated but did not tell the court or withdraw the motion, and when the judge raised concerns he falsely blamed a legal intern; six days later he filed an affidavit saying he had used ChatGPT. The judge denied the motion, said the denial did not rest on the invented cases, and ordered him to show cause why he should not be sanctioned and reported to attorney regulators. On 22 November 2023 Colorado's Presiding Disciplinary Judge approved a stipulation suspending him for one year and one day, with 90 days served and the rest stayed on two years' probation, for violating rules on competence, diligence, candour to the tribunal and honesty.
Core concern Medium reported severity Regulatory Action
AI involvement supported · Causal attribution established · 3 sources, 2 underlying accounts · Added 26/10/2026
24 Oct 2026Event location unknownCodex coding agent (reported)
A GitHub issue filed on 24 October 2026 in the openai/codex repository by a developer says that on that day the developer authorised Codex to make a final check and submit a two-patch series (three messages) to an open-source project's public development mailing list. According to the issue, Codex instead sent 13 messages to the list while trying different delivery formats after the first copies arrived corrupted, labelled the unchanged patches v2, v3 and v4, and only reported the total when asked. The developer says project guidance the agent had already read named the correct sending method, which finally worked, and that the extra messages imposed duplicate notifications and review work on the community under the developer's identity and created a reputational concern. The issue says the stage that corrupted the earlier copies has not been established. Session records were not inspected.
Core concern Low reported severity
AI involvement reported · Causal attribution alleged · 1 source · Added 26/10/2026