Sept 2026NetherlandsUnidentified voice-generation tool
The Dutch police reported on 21 October 2026 that the Rotterdam Team Cybercrime had stopped a network behind a new form of bank-helpdesk fraud. According to the police, the fraudsters had used artificial intelligence to make dozens of audio fragments with different voices, banks, companies and scenarios. Victims were called and heard a fragment warning of suspicious activity on their account; pressing 1 put them through to a fake bank employee, who had them install an app and hold their bank card to their phone, after which an accomplice withdrew money at a cash machine. In the three weeks the team could see the network's servers, the police counted about 170 victims of all ages and about 380,000 euros in losses. Two suspects were arrested on 6 October 2026 and a Rotterdam court extended their detention by at least 90 days on 20 October.
Core concern Medium reported severity Criminal Charges
AI involvement reported · Causal attribution alleged · 2 sources, 1 underlying account · Added 22/10/2026
Sept 2026Event location unknownUnidentified image tool
In mid-September 2026 two versions of an image that appeared to show US President Donald Trump kissing a woman on a golf course spread on social media, where posters presumed the woman was a named White House aide and treated the scene as evidence of an intimate relationship. Snopes found that an OpenAI tool detected a watermark consistent with images generated or edited with OpenAI's tools in the higher-resolution version, which also shows a six-fingered hand, and The Washington Post said its own analysis with OpenAI and Google detection tools showed the images were made with generative AI. A White House official called the image clearly a fake photo. Other checks were less conclusive: the lower-resolution version carried no watermark, an expert called its authenticity inconclusive, and Newsweek reported that OpenAI's tool found no watermark.
Core concern Low reported severity Media Coverage
AI involvement reported · Causal attribution alleged · 4 sources · Added 22/10/2026
21 Oct 2026Event location unknownClaude Code (reported)
A GitHub issue opened on 22 October 2026 (UTC) by a developer says that a background Claude Code subagent, running a script it had written through the Bash tool on the developer's Ubuntu Linux machine, ended up calling kill(-1, SIGKILL), which killed every process the developer's user account owned: the tmux server, every other Claude Code session on the machine, the desktop shell and Chrome. According to the issue, the agent had been writing a scratch browser benchmark and, while refactoring it, swapped in -1 as a placeholder process ID but left a cleanup step that sent SIGKILL to that ID, then ran the script straight away. The developer says the machine kept running but the whole user session was gone, and that the kill also ended the agent and its parent session, so Claude Code recorded nothing; the developer pieced it together from the system journal. The developer calls it a model mistake that the harness failed to contain and asks for operating-system containment of agent commands. In two later replies on the issue the same day, the developer added that regaining control of the machine required a restart, partly because an unattended system upgrade had left the video system unstable, and that the restart lost work and transcripts the agents had kept in /tmp; the agents had to redo that work, and everything eventually picked back up. Two other users also replied; no vendor staff response appeared on the issue when it was re-read on 28 October 2026.
Core concern Low reported severity
AI involvement reported · Causal attribution alleged · 1 source · Added 22/10/2026
14 Oct 2026South KoreaUnidentified AI problem-solving tool
In a midterm exam on 14 October 2026 in a major course of about 100 students at Konkuk University in Seoul, a student sent the questions to a friend through a camera attached to the chest; the friend solved them using AI and sent the answers to the student's smartwatch, according to the professor's notice to students as reported by JoongAng Ilbo on 21 October 2026. The professor became suspicious of a perfect score, asked the student to solve the same problems again, and the student could not and confessed. The university said the professor gave the student a zero on the exam and an F in the course under its regulations for cheating by writing answers using AI and a smartwatch, and the department decided to ban cameras and watches from the final exam onward. The AI tool is not named, and no consequence for the friend is reported.
AI relation unknown Low reported severity Internal Action
AI involvement reported · Causal attribution supported · 2 sources, 1 underlying account · Added 22/10/2026
11 Feb 2026Event location unknownUnidentified image tool
Tokyo's Metropolitan Police Department arrested a 32-year-old company employee on 1 August 2026 and referred a 17-year-old high-school student to prosecutors, on suspicion of defamation and public display of child pornography. According to the police, on 11 February 2026 the two, acting together, used generative AI to turn a photo of a woman in her 40s, taken in her first year of junior high school in gym clothes, into two fake nude images, and posted them to a group on social media; the student had asked the man to make the edits. Police say the man is also suspected of posting similar edits of three women in their 20s, that neither suspect knew the women, and that about 3,800 obscene images and videos were found on his phone. The Yomiuri Shimbun, citing investigative sources, reports that police believe he made about 300 such images of women in athletics uniforms or gym clothes and posted about 100, about half at the student's request, and that he admitted the allegations.
Core concern Medium reported severity Criminal Charges
AI involvement reported · Causal attribution alleged · 2 sources · Added 21/10/2026
May 2026 to Jun 2026Event location unknownUnidentified image and voice tool
The Oberbergischer Kreis police reported on 21 July 2026 that a woman had been in contact since mid-May with an unknown person posing as a widowed doctor from Ukraine, who exchanged chats, pictures and voice messages with her through a messaging service. On 23 June she transferred a mid four-figure sum the supposed doctor requested; bank staff held back a second, high four-figure transfer at the end of June. She reported the fraud on 20 July, and the police say first investigations showed that the pictures and voice messages she had received were apparently created with artificial intelligence.
Core concern Medium reported severity Investigation Opened
AI involvement reported · Causal attribution alleged · 1 source · Added 20/10/2026
Oct 2026IndiaUnidentified Google AI tool (reported)
In October 2026 a ticket inspector on the Somnath/Veraval-Jabalpur Express, after the train left Wankaner in Gujarat, found that two passengers were carrying fake tickets on their mobile phones; they were handed to the Railway Protection Force and railway police in Rajkot registered a case and detained them. According to Gujarat Samachar, the two were factory workers, one of them a minor, who wanted to go home and asked a man working at their factory for tickets; he made the tickets with the help of AI and Photoshop and charged Rs 2,000. Gujarat Samachar English reports that the railway police investigation traced the forged tickets to a Google AI application and WhatsApp, and that the man who made them was also arrested. The reports differ on the workers' role: Bombay Samachar says they were cheated and at first suspected nothing, while Gujarat Samachar English says all three accused, including the two passengers, were remanded to judicial custody. Gujarat Samachar English dates the check 15 October, and the reports name no specific AI product.
Core concern Medium reported severity Criminal Charges
AI involvement reported · Causal attribution supported · 3 sources, 2 underlying accounts · Added 20/10/2026
Sept 2026Event location unknownBooking.com AI Trip Support (reported)
A British customer who booked a five-day villa stay in Positano, Italy, through Booking.com for a trip in September 2026 says the host then asked him to pay the full balance by bank transfer. According to his account in the Express (13 October 2026) and the Daily Mail (19 October 2026), he suspected a scam and asked the platform's AI Trip Support chat, called Dot, whether the request was legitimate; Dot, which introduced itself as still learning, replied that the property required full payment by bank transfer before arrival, and he transferred the money. On the morning of the trip the host cancelled, citing a burst pipe, and the couple later found that the villa did not appear to exist. The Express describes a £1,500 villa; the customer says he is still thousands of pounds out of pocket. Booking.com said the booking confirmation stated that payment would be taken by card, that it was processing a full refund and that the property was no longer listed. The chatbot exchange rests on the customer's account; Booking.com's quoted statement does not address it.
Core concern Low reported severity Media Coverage
AI involvement reported · Causal attribution alleged · 2 sources, 1 underlying account · Added 20/10/2026
16 Feb 2026Event location unknownYuanbao
A lawyer told Huashang Daily that on the evening of Chinese New Year's Eve 2026 he uploaded his photo to Tencent's Yuanbao app and asked it, over about five instructions, to design a New Year greeting image suited to his profession. He said he used no banned or leading words. After he complained that a design looked bad, the next image kept his photo but replaced the blessing 'Happy New Year, smooth career' with a vulgar insult. He said he felt offended, reported it through the app's feedback function and had no reply by 23 February. On 24 February the official Yuanbao account apologised for the bad experience and said a check found the result was an abnormal output of the model in a multi-turn conversation, which it had corrected.
Core concern Low reported severity Internal Action
AI involvement supported · Causal attribution supported · 2 sources · Added 19/10/2026
2026ChinaUnidentified video tool
The Beijing News reported on 18 October 2026, in a follow-up to its investigation of 'Xiaohai' accounts on Chinese short-video platforms, that the accounts show AI-generated teenage boys who address older women as 'aunties and mothers' and steer them to high-priced shop windows. A university student told the paper that an older relative of a friend had been watching 'Xiaohai' videos and asked the friend why the boy was so badly off; the friend then found that the relative had already bought several orders of goods at high prices through the persona's shop window. The student made a counter-video with the same AI tools, posted on 28 September, in which the persona admits to deceiving viewers; after the friend showed it to the relative, the relative said they would not buy the persona's goods again. The amount spent and the goods bought are not reported. China National Radio's report the same day also describes the 'Xiaohai' videos as AI-generated and notes that many carry an 'AI-generated' watermark.
Core concern Low reported severity Media Coverage
AI involvement reported · Causal attribution alleged · 2 sources · Added 19/10/2026
14 Jul 2026ChinaUnidentified image tool
China Daily reported on 15 July 2026 that the first prize in a local photography competition in Hohhot, Inner Mongolia, had been revoked after the image was found to be AI-generated. Online users had pointed to garbled characters on the workers' vests, unnatural sunlight and matching faces in the winning photo of three sanitation workers. After an investigation, the city's federation of literary and art circles, one of the organisers, confirmed in a notice on Tuesday 14 July that the photo had been created using AI, cancelled its award and barred the work from competitions, evaluations, exhibitions and official archives. The competition was suspended for rectification; the notice said those responsible would be dealt with under the relevant rules and that the case exposed lax vetting and imprecise judging criteria. The entrant is not named, and who generated the image is not stated.
AI relation unknown Low reported severity Internal Action
AI involvement reported · Causal attribution supported · 2 sources, 1 underlying account · Added 19/10/2026
2026Event location unknownUnidentified AI system
In a first-person essay published by the Oakland Review of Books on 2 October 2026, a former customer success manager at an education technology company writes that the company began a push at the end of 2025 to become 'AI-first', requiring staff to use AI tools, and that the author was later laid off. The author writes that an hour after the layoff the CEO held a seven-minute all-hands meeting to read a Claude-generated script saying leadership was 'replacing these roles with AI', and that the company let 40 people go, about 20 percent of its workforce, at the end of about six months of role eliminations. Staff who could not attend the HR meeting were told by email that their roles ended the next business day. The author also writes that sales were down and customers were unhappy, and that losing the job brought more relief than anything else, cushioned by savings and an employed partner. The company is not named, and the layoff is not dated beyond the essay's sequence. The account is uncorroborated.
AI relation unknown Medium reported severity
AI involvement reported · Causal attribution alleged · 1 source · Added 19/10/2026
Sept 2026Event location unknownMeta Muse agent
In an as-told-to essay published by Business Insider on 15 October 2026, a software developer says that three days after installing Meta's Muse agent, and two days after last using it, the agent sent an unprompted message saying it could not check the developer's email for an 'Anthropic message' because Gmail was not connected. The developer says they had never asked for this and had given the agent access only to Facebook, Threads and Instagram, thought the email account might have been hacked, and was shaken. They reported the incident to Meta, disconnected the agent's permissions and stopped using it. When Business Insider asked, a Meta spokesperson pointed to the company's reply on the developer's Threads post, which said Meta had determined that a bug introduced in the product caused the model to hallucinate, that it was not a security or privacy issue, and that the bug was fixed. The developer says no damage was done. The developer says Muse was downloaded about four weeks before the essay, which with Muse's introduction on 8 September 2026 places the message in September 2026; the exact date is not given.
Core concern Low reported severity Internal Action
AI involvement supported · Causal attribution supported · 2 sources · Added 19/10/2026
2026Event location unknownUnidentified AI model
Times of Malta reported on 18 October 2026, from interviews with anonymous former gambling-sector employees, that iGaming companies are laying off staff as they turn to AI. One former designer, who says they lost their job earlier in 2026, says their team had been told to train an AI model to make design and brand-guideline decisions, presented as a tool to help designers, and that the company later decided the model was fit enough to replace around half of the design team, including those who trained it. The designer says there was no notice, that they were asked to resign later that day and that the severance was the bare minimum. The outlet omitted company names, and the workers said they had signed non-disclosure agreements. GamingMalta, the industry promotion foundation, told the outlet it sees talent being redeployed rather than replaced by AI, and the article notes that others pointed to search-engine changes and higher UK gambling taxes. The account is a single anonymous worker's, relayed by one outlet; the employer and where the job was based are not stated.
AI relation unknown Medium reported severity
AI involvement reported · Causal attribution alleged · 1 source · Added 19/10/2026
Mar 2026JapanUnidentified video tool (suspected)
The Yomiuri Shimbun reported on 19 March 2026 that in early March an applicant in a Tokyo IT company's online mid-career job interview used the name, career history and apparent face of a real engineer who heads a Tokyo software company. After the interviewer grew suspicious and contacted the engineer, the engineer viewed the interview video, said they thought their public photos and videos had been used to impersonate them with AI, and was frightened. Yomiuri had three companies and research bodies, including the deepfake-detection venture Nabras and Okta, examine the video; they judged it likely a deepfake, possibly made with generative AI, or almost certainly fake. Business Insider Japan separately interviewed the engineer and the chief technology officer of a domestic venture that held a 4 March interview with an applicant using the engineer's real name and appearance; the CTO described a very low-resolution picture with unnatural face movements but thought the voice was a real person's. Who the applicant was is not established.
Core concern Low reported severity Media Coverage
AI involvement suspected · Causal attribution alleged · 2 sources · Added 19/10/2026
May 2026ChinaUnidentified video tool
The Beijing News reported on 16 October 2026, in an investigation of 'Xiaohai' accounts on Chinese short-video platforms, that the accounts show an AI-generated digital teenager who calls older women viewers 'aunties and mothers', tells stories of hardship and asks them to buy goods through sales links; its reporter generated such a video by following a tutorial, in a little over ten minutes. The paper describes a 79-year-old woman, given a pseudonym, who lived alone and talked most with the 'Xiaohai' on her phone. Around May 2026, according to the paper, a 'Xiaohai' video invited viewers to Yuncheng in Shanxi to see apricot blossoms; she secretly packed, bought a train ticket and set off with no clear route. On the train, railway police noticed her, told her that 'Xiaohai' was not a real person, and contacted her children, who bought her return ticket and brought her home. At first she refused to believe it. The paper also found that most 'Xiaohai' videos now carry a declaration that the content is AI-generated. How the paper learned of her journey is not stated. China National Radio (China Media Group) reported the same journey on 19 October 2026 with recorded exchanges between the woman and a train police officer: in May 2026 Urumqi railway police found a 79-year-old who cannot read travelling alone, able to say only that she was going to Yuncheng in Shanxi to see a child who called her 'auntie' every day on her phone; the officer told her the person in the live room was fake, and she got off and went home on a return ticket her family bought. The station's reporter found that the persona she trusted was one of the batch-produced 'digital human Xiaohai' accounts and that many carry an 'AI-generated' watermark in a corner of the video.
Core concern Low reported severity Media Coverage
AI involvement supported · Causal attribution alleged · 3 sources, 2 underlying accounts · Added 18/10/2026
26 Mar 2026 to 19 Aug 2026AustraliaUnidentified AI drafting tool
On 19 August 2026 a Deputy President of Australia's Fair Work Commission ordered a self-represented former Aldi store worker to pay A$1,230 towards Aldi's legal costs. The decision records that his unfair dismissal application, lodged on the evening he was told of his dismissal on 26 March 2026, was obviously prepared using AI and still contained AI drafting notes, and that every email and document he filed was drafted using AI. He was notified of his dismissal three days short of the six-month minimum employment period. According to the decision, his AI-generated materials kept arguing about a different, irrelevant date (when the dismissal took effect), including after the Commission explained the correct test to him twice by email on 31 July 2026. He discontinued at the hearing on 5 August 2026. The Deputy President found that continuing after those warnings was unreasonable and said it was quite possible he had adopted AI as his quasi-legal adviser, while stating that using AI to draft submissions is not of itself unreasonable and that the responsibility was his. The decision notes that an appeal was lodged against it.
Core concern Low reported severity Regulatory Action
AI involvement supported · Causal attribution supported · 2 sources, 1 underlying account · Added 18/10/2026
15 Jul 2026United StatesUnidentified AI drafting tool
In a published opinion of July 2026, the Arizona Court of Appeals found that a self-represented appellant's opening brief in a probate dispute contained eight citations, six of them deficient, including two cases that do not exist. In a written explanation and at a hearing on the false citations he acknowledged relying on generative AI to write the brief and said he did nothing to verify the citations. The court struck the inaccurate and false citations, rejected the appeal because no supportive authority remained, held that citing a hallucinated case is sanctionable whether or not it is intended, and ordered him to pay the share of the other party's attorney's fees traceable to the fake citations. Because he said he did not intend to deceive and is not legally trained, the court chose this narrow sanction over contempt or fines. It described the opinion as the first published Arizona ruling on generative AI hallucinations in filings.
Core + contextual relations Low reported severity Regulatory Action
AI involvement supported · Causal attribution established · 1 source · Added 18/10/2026
May 2026Event location unknownChatGPT and Claude (reported)
Ars Technica reported on 22 May 2026 that a New York Times investigation that week found what Steven Rosenbaum acknowledged are "a handful of improperly attributed or synthetic quotes" in his book The Future of Truth: How AI Reshapes Reality, linked to his use of AI tools while researching it. Rosenbaum told Ars that ChatGPT and Claude were used heavily to gather information and that AI-sourced material was tagged in his notes and passed to a fact-checker and copy editors provided by the publisher. According to Ars, the Times identified six problematic citations among the book's 285 outside citations, including three synthetic quotes with no apparent source. Tech reporter Kara Swisher told the Times she never said a quote attributed to her, and professor Lisa Feldman Barrett said quotes attributed to Barrett do not appear in Barrett's book and are wrong. Rosenbaum said he was working with editors on a citation audit to correct future editions. WIRED later removed an excerpt from the book, saying it did not meet its editorial standards.
Core + contextual relations Low reported severity Media Coverage
AI involvement reported · Causal attribution supported · 2 sources · Added 18/10/2026
6 May 2026United StatesClaude Console (reported)
In the federal lawsuit over the Trump administration's mass layoffs (AFGE v. Trump, Northern District of California), a partner at Binnall Law Group filed an emergency motion on the night of 6 May 2026, in the Eastern District of Virginia, to quash a deposition subpoena directed at a former Homeland Security official the firm represents; that court transferred the motion to the Northern District of California. On 13 May Judge Susan Illston denied the motion. Her order says the court was troubled by multiple citations to case quotations that do not exist in the cases cited, that counsel was unaware of these quotations when questioned at the hearing and had no explanation for them, and it ordered counsel to explain the fabricated quotations in a declaration under penalty of perjury by 15 May. In that declaration the partner said he had used Claude Console, an enterprise AI platform, to create an initial draft, that an associate told him she had checked the citations, and that he did not double-check the quotations. He apologised to the court and opposing counsel and attached a corrected version. The firm's founding partner declared that firm policy prohibits using tools like Claude for legal drafting, that the error breached its policies and training, and that the firm had opened an internal investigation. On 20 Oct 2026 the court administratively closed the related miscellaneous case, saying the motion had been adjudicated and no further filings were anticipated. Neither order read mentions a sanction, and whether one was imposed is not established.
Core concern Low reported severity Internal Action
AI involvement supported · Causal attribution supported · 4 sources, 3 underlying accounts · Added 18/10/2026