Event date unknownUnited KingdomAnima Annie
Fortune reported on 20 July 2025 that a London patient, a healthy man in his mid-20s, had recently received a letter inviting him to diabetic eye screening although he had never been diagnosed with diabetes. He told Fortune he briefly worried he had been diagnosed without knowing, then assumed an admin error. The next day a nurse at a routine blood test questioned the diagnosis and showed him the notes on the system, which were AI-generated summaries. In his full records, which Fortune reviewed, the entry that introduced the diagnosis was listed as a summary 'generated by Annie AI', the tool of Anima Health. It said he had presented with chest pain and shortness of breath from likely angina due to coronary artery disease, that he had been diagnosed with type 2 diabetes and was on medications with dosages, and it gave the source document's address as a fictitious 'Health Hospital' on '456 Care Road' in 'Health City'. According to the patient and other records Fortune reviewed, none of these details was accurate; the entry appeared around the time he had attended hospital with tonsillitis, which it did not mention. Fortune writes that the invitation was presumably based on the erroneous summary. An NHS representative told Fortune the practice makes limited use of supervised AI, that Anima only suggests codes and a summary to a human reviewer, and that the error was a one-off human error: a summariser had spotted the mistake but, distracted, saved the original version instead of the corrected one. Anima Health did not respond to Fortune.
Contextual tracker case Low reported severity Media Coverage
AI involvement reported · Causal attribution alleged · 1 source · Added 26/10/2026
May 2026ChinaDoubao (reported)
Nanguo Zaobao, a Nanning daily, reported on 27 May 2026 that first-time parents in Nanning, Guangxi, had fed their one-month-old baby strictly 60 ml of milk per feed for about a week because, by their account, the chatbot Doubao had advised that amount. The baby cried constantly and slept only when exhausted. At a jaundice recheck in late May the hospital's neonatology director said a full-term one-month-old normally takes 80 to 100 ml per feed and that the baby had not gained even half a kilogram. On 28 May Doubao called the report untrue, saying its own repeated tests did not produce that advice. The conversation itself has not been published.
Contextual tracker case Medium reported severity Involving minors Media Coverage
AI involvement disputed · Causal attribution disputed · 3 sources, 2 underlying accounts · Added 26/10/2026
1 Dec 2025CzechiaUnidentified legal research tool
On 1 December 2025 the presiding judge of a chamber of the Czech Constitutional Court fined a Prague lawyer 25,000 crowns (a quarter of the statutory maximum) for grossly obstructing proceedings. The lawyer had filed a constitutional complaint, for a client refused a temporary residence permit, that argued from 12 decisions of the Constitutional Court and the European Court of Human Rights; the court found that a substantial part of them did not exist and part were grossly misinterpreted. The order says the complaint showed signs of generative AI and that the court presumed the lawyer had generated it with an AI model and then failed to verify the citations. The lawyer confirmed to INFO.CZ that an AI tool was used, described it as a paid professional model from a legal information system provider, said the output was not checked enough, and expressed regret. The court invited the client to correct the complaint so that the client would suffer no procedural harm; the complaint was later rejected as manifestly unfounded on 30 January 2026. INFO.CZ reported on 10 December 2025 that the lawyer had apologised and paid the fine and that the Czech Bar Association's supervisory board was examining the case.
Core + contextual relations Low reported severity Regulatory Action
AI involvement supported · Causal attribution established · 5 sources · Added 26/10/2026
19 Aug 2025IndiaUnidentified AI tool
On 19 August 2025 a trial court in Andhra Pradesh hearing a suit for an injunction over a property dismissed the defendants' objections to an advocate-commissioner's report on the property, relying on four purported Supreme Court judgments that, according to the defendants and later the courts, do not exist. The defendants challenged the order in the Andhra Pradesh High Court, which, as the Supreme Court recorded, realised the judgments were AI-generated, recorded a word of caution, and on 21 January 2026 dismissed their petition on the merits. On appeal, the Supreme Court said in an order of 27 February 2026 that it took cognisance of the trial court deploying AI-generated non-existing, fake or synthetic alleged judgments, that a decision based on such judgments is not an error in decision making but misconduct, and directed that the trial court not proceed on the basis of the advocate-commissioner's report while the appeal is pending. In May 2026 the Supreme Court asked the Bar Council of India to set up an expert panel on AI-generated fake judgments. The AI tool is not named, the sources read do not name the judge, and no outcome of the appeal had been reported in the sources read.
Contextual tracker case Low reported severity Internal Action
AI involvement reported · Causal attribution supported · 4 sources, 2 underlying accounts · Added 25/10/2026
2024GermanyMicrosoft Copilot
A court reporter from the Tübingen area found in 2024 that Microsoft's Copilot, asked about him, described him as a man charged in a child abuse case, an escapee from a psychiatric institution and a fraudster who cheated widows, according to his account to the public broadcaster SWR as relayed by heise and ComputerBild. All were defendants in trials he had covered as a journalist. heise and ComputerBild, both citing SWR, report that Copilot also gave his private phone number and full home address and offered a route there. He told The Register that seeing his name tied to these crimes was traumatising and that prosecutors had rejected his criminal complaint in two instances; according to the SWR account relayed by heise, the complaint failed because no real person was the author. After the Bavarian data-protection authority contacted Microsoft the claims were briefly unavailable and then returned. Microsoft told The Register it had investigated and taken appropriate and immediate action.
Contextual tracker case Medium reported severity Internal Action
AI involvement supported · Causal attribution supported · 4 sources, 3 underlying accounts · Added 25/10/2026
23 Oct 2026SpainPangram AI-text detector
El País checked the winning novel of the first Almadía Ventosa-Arrufat short-novel prize (20,000 euros; the rules banned generative AI) with the Pangram AI-text detector, which scored 99% of the roughly 26,000-word text as AI-generated, according to outlets relaying the investigation. El País passed its suspicions to the Mexican publisher Almadía, which on Friday 23 October 2026 announced a detailed review, saying it would not anticipate conclusions, and suspended all activities presenting the book, including a planned tour in Madrid, Barcelona, Seville and Cádiz. The publisher said the Peruvian author had declared in his contract and personally that he did not use AI, and that neither the jury nor its editors had seen signs of AI use. In a statement on Instagram reported by La Nación on 29 October 2026, the author said he conceived, wrote and revised the novel without generative AI tools and was gathering earlier manuscript versions, his notebook and research materials to give to the publisher for its review. Writers objected that a detector score is not conclusive proof of authorship. Whether AI was used is unestablished.
Contextual tracker case Low reported severity Investigation Opened
AI involvement reported · Causal attribution supported · 5 sources, 4 underlying accounts · Added 25/10/2026
Apr 2025NetherlandsChatGPT
The Rechtbank Overijssel (Zwolle location) ordered on 5 January 2026 that a couple's marriage record of April 2025 be struck out of the Zwolle register of marriages, holding that no marriage had come into existence. The couple had asked an acquaintance to conduct the ceremony as a one-day special registrar, and because they wanted a light-hearted ceremony the officiant wrote the speech with the help of ChatGPT. The speech, which the court quotes, asked the partners to stand by each other and declared them 'a team', but it did not contain the declaration that Dutch civil law (article 1:67(1) of the Civil Code) requires spouses to make. The public prosecution service says it brought the case after a report from the municipality. The couple said the error was not their responsibility, that the municipal registrar present had not corrected it, and that losing their wedding date had a great emotional impact; the court refused their request to keep the original date.
Contextual tracker case Low reported severity Lawsuit Dismissed
AI involvement supported · Causal attribution supported · 2 sources, 1 underlying account · Added 25/10/2026
20 Oct 2026Event location unknownUnidentified video tool (suspected)
On 20 October 2026, during early voting in Texas, a newly created X account posted two videos in which a man presented himself as a former student of James Talarico, the Democratic US Senate candidate and a former middle-school teacher, and accused him of sexual assault in 2012. The threat-intelligence firm Graphika said the account very likely impersonated a real Facebook user of the same name who had been Talarico's student and had praised him, assessed the videos as very likely deepfakes because the face moved unnaturally or out of sync with the audio and AI analysis tools indicated AI generation, and attributed them with high confidence to the Russia-aligned influence operation Storm-1516. PolitiFact reported that NewsGuard found many signs of AI generation and a 98 percent detector score, and that the real former student told the fact-checkers Lead Stories the video was not of him and denied accusing Talarico. Talarico's campaign said he was being slandered with AI deepfakes, foreign bots and conspiracy theories. PolitiFact rated the claim Pants on Fire. Who made the videos and how many people saw them is not established.
Core + contextual relations Medium reported severity Media Coverage
AI involvement suspected · Causal attribution supported · 2 sources · Added 25/10/2026
2025RussiaAntiplagiat AI-text detector
RAPSI and BIZNES Online reported on 8 and 10 June 2026 that the Supreme Court of Tatarstan upheld a ruling that a private Moscow university unlawfully expelled a distance-learning student in 2025 over suspected use of AI in her graduation thesis. According to the university's representatives in court as reported, the Antiplagiat.Vuz system first showed about 41% (41.91%) AI-generated text, her supervisor then said the word order had been deliberately disturbed in more than 70% of the text, and a third check showed 11.92%. BIZNES Online also reports, from the court documents, that a university representative told the court the supervisor does not operate the checking system and had not in fact analysed the text for neural-network generation, and could not explain how the system distinguishes borrowed text from AI generation. The university refused the revised work as late (uploaded on Monday 9 June 2025 against a deadline of 8 June) and expelled her; the department's protocol cited a breach of good-faith work in the Antiplagiat.Vuz system. She denied using AI. In January 2026 the Vakhitovsky District Court of Kazan held the expulsion unlawful, ordered her reinstatement for a check of her thesis and, if it met the originality threshold, admission to the defence, and awarded 40,000 roubles in moral damages and a 20,000-rouble fine. The Supreme Court of Tatarstan dismissed the university's appeal: it held that the probabilistic report had not been checked by a competent expert and could not be the sole proof of bad faith, noted that the same text was classified as AI-generated in one report and not in another, and found no evidence that any flagged fragments resulted from her deliberate acts rather than her sources or an incorrect algorithm. The courts also held the deadline had been miscounted.
Contextual tracker case Medium reported severity Lawsuit Filed
AI involvement supported · Causal attribution supported · 3 sources, 1 underlying account · Added 25/10/2026
May 2024ItalyUnidentified speech-to-text tool
In May 2024 a businessman questioned as a suspect in the Genoa prosecutors' inquiry into alleged bribes to the president of the Liguria region was recorded in the official transcript of his interrogation before the preliminary-investigations judge as saying that the president wanted 'finanziamenti illeciti' (unlawful financing). According to Corriere della Sera, on receiving the transcript he wrote to the judge to contest it, saying he had said 'finanziamenti leciti' (lawful financing); his lawyer said the legally valid handwritten record did not contain the word. Il Post reports that the passage had been highlighted in the headlines of many newspapers in the following days as confirming the prosecutors' accusations, and that a second hearing of the audio was needed to settle it. La Repubblica reported on 27 May 2024 that, at an ad hoc hearing, the businessman confirmed he had said 'leciti', with the prosecutor's agreement according to his lawyer, and the judge took note; it wrote that the fault appears to lie with the automatic software that transcribes words spoken in court. Il Post reports that checks found the error was not human and that the transcript had been produced by speech-to-text software. The software is not named.
Contextual tracker case Low reported severity Internal Action
AI involvement reported · Causal attribution supported · 3 sources · Added 25/10/2026
7 Jun 2024NetherlandsChatGPT
A homeowner in the Netherlands sued his neighbour, whose permitted roof extension shaded his solar panels, claiming more than 21,000 euros for lost yield and investment. In a judgment of 7 June 2024 (ECLI:NL:RBGEL:2024:3636), the sub-district judge of the Rechtbank Gelderland, sitting in Nijmegen, held that the neighbour had to bear half of the damage and then calculated the damage, estimating the average lifespan of solar panels from 2009 and the current average electricity price partly with the help of ChatGPT, as the judgment states. Using a lifespan of 27.5 years, about 15 remaining years and a price of 0.30 euros per kWh, the judge put the yield loss at about 2,250 euros and the maximum damage at 3,323.40 euros; half of that was less than the 3,300 euros the neighbour had offered, so the claim was rejected and the homeowner, who had refused the offer, received nothing and was ordered to pay costs. On the neighbour's counterclaim, the judge assumed, again partly with ChatGPT, that insulation material the homeowner had thrown into water was no longer usable, and ordered him to pay damages of 2,619.02 euros in total. The judgment does not say what was asked or answered, and whether it was appealed is not reported.
Contextual tracker case Low reported severity Lawsuit Dismissed
AI involvement supported · Causal attribution supported · 2 sources, 1 underlying account · Added 24/10/2026
2021IndiaSamagra Vedika
An Al Jazeera investigation published on 24 January 2024, produced with The Reporters' Collective and the Pulitzer Center's AI Accountability Network, reports that Samagra Vedika, a system the Telangana government uses to consolidate residents' data and check welfare eligibility, wrongly tagged poor families in Hyderabad as owning cars, and that officials relied on those tags to deny them food security cards (subsidised rice). Al Jazeera says the system uses machine learning and entity resolution to match people across databases, and that since 2016 officials have been mandated to check whether the system approved an applicant before deciding. In one case it reports that the system mistook a widow's late husband, a rickshaw puller, for a car owner with a similar name; her renewal application was rejected in 2021 for possessing a four-wheeler. After she traced the vehicle's real owner, officials agreed there had been an algorithmic error but said her family income exceeded the limit, which Al Jazeera says was not the case; she had been without the subsidised rations for more than seven years. A second family, whose husband cannot drive, was rejected in 2021 for possessing a four-wheeler; in November 2023 the High Court of Telangana said the petitioner is eligible whenever new cards are issued. The government says the technology is highly precise and that wrong vehicle or house tags occur in under five per cent of cases and have mostly been corrected.
Contextual tracker case Medium reported severity Media Coverage
AI involvement supported · Causal attribution supported · 1 source · Added 24/10/2026
Event date unknownEvent location unknownUnidentified chatbot
Panasonic Operational Excellence, a Panasonic Holdings subsidiary, announced on 23 October 2026 that a member of an in-house project introducing a personnel information system had uploaded part of that system's data, as registered in April 2026, to an external generative AI service whose use the company prohibits. According to the announcement as reported by ITmedia and CNET Japan, the data covered about 75,000 current and about 178,000 former employees of Panasonic group companies in Japan (employee numbers, addresses, phone numbers and email addresses, without names) and about 81,000 family members registered as emergency contacts (names, addresses, phone numbers and email addresses), about 334,000 people in all. The company says the upload broke internal rules and was not an outside attack, that the employee deleted the chat history and the local data immediately after the problem came to light, and that it reported the matter to Japan's Personal Information Protection Commission on 29 September. The provider told the company that training on the data was switched off at upload but that deleted data is kept for 30 days, until 25 October. No public exposure or misuse had been confirmed. The company has not named the AI service.
Contextual tracker case Medium reported severity
AI involvement reported · Causal attribution supported · 3 sources, 1 underlying account · Added 24/10/2026
Dec 2023 to Mar 2024United StatesGrammarly (reported)
In late 2023 a junior at a public high school in Hingham, Massachusetts, and a classmate submitted a script for an AP U.S. History project that, a federal court later found on a preliminary record, contained text copied from Grammarly, a publicly available AI tool, including citations to nonexistent books, with no acknowledgement of AI use. Turnitin and other AI detection tools had flagged the script, and a school official says the pupil admitted using Grammarly. In December 2023 the school treated it as cheating: the pupil failed two parts of the project, served a Saturday detention and was rejected from the National Honor Society in spring 2024, though he was later admitted. His parents sued, arguing the school had not made its AI rules clear; on 20 November 2024 the US District Court for the District of Massachusetts refused a preliminary injunction to undo the penalties, and in February 2025 the parties dismissed the case with prejudice.
Core + contextual relations Low reported severity Involving minors Lawsuit Dismissed
AI involvement reported · Causal attribution supported · 2 sources · Added 23/10/2026
Event date unknownUkraineChatGPT (reported)
The Appeals Chamber of Ukraine's High Anti-Corruption Court left without consideration an appeal that the applicant had drafted with the ChatGPT chatbot, treating the approach as a sign of disrespect for the judicial system, UkraNews reported on 15 June 2025 from a ruling in the court-decision register. According to the report, the applicant wrote in the appeal that she had turned to AI after her lawyers declined to prepare it, and the court said the text contained sarcastic and offensive wording about the courts and that using ChatGPT-generated material without proper legal grounds contradicted the principles of judicial proceedings. The appeal concerned an investigating judge's refusal of her complaint about inaction by anti-corruption investigators and prosecutors. The ruling could be challenged in cassation within three months; no later outcome was found.
Core + contextual relations Low reported severity Lawsuit Dismissed
AI involvement reported · Causal attribution supported · 1 source · Added 23/10/2026
Event date unknownRussiaUnidentified legal research tool
Russia's Court for Intellectual Property Rights fined an individual entrepreneur from Udmurtia 5,000 roubles for contempt of court after finding that part of the 12 references to case law in his cassation appeal, including one said to be a Supreme Court position, pointed to court acts that do not exist or attributed conclusions to them, the court's press service told RIA Novosti on 27 August 2026. Rossiyskaya Gazeta reports that the entrepreneur explained he had prepared the appeal with an AI-based service that selected the case law, and that his side argued he had no legal education and could not check the references; the court rejected the argument that the errors were technical. The date of the ruling and the AI service are not reported.
Core + contextual relations Low reported severity Regulatory Action
AI involvement reported · Causal attribution supported · 2 sources · Added 23/10/2026
Event date unknownEvent location unknownUnidentified AI-text detector
A case summary published in July 2025 by the Office of the Independent Adjudicator for Higher Education (OIA), the independent student complaints scheme for England and Wales, describes a student with autism who was accused of including large sections of AI-generated content in an essay. The student denied using AI and submitted planning and notes. A disciplinary panel, which had the provider's AI detection software report and compared the essay with other work by the student, found academic misconduct and gave a mark of zero with any reassessment capped at a pass; the provider rejected the appeal, in which the student said the same software had earlier flagged another of their essays in a case where it had been agreed they had not used AI. The OIA found the complaint Justified: the panel had not properly considered the student's evidence, had not shared some evidence it relied on, and had not given sufficient reasons. It recommended reconsideration and changes to the provider's rules, and the student told the OIA that on reconsideration the provider decided there had been no misconduct. The provider and the software are not named.
Contextual tracker case Low reported severity Regulatory Action
AI involvement reported · Causal attribution supported · 2 sources · Added 23/10/2026
9 Dec 2025United StatesZeroEyes gun detection
On 9 December 2025 a Seminole County, Florida, middle school went into a Code Red lockdown and police rushed to the campus after the ZeroEyes AI weapons detection system, whose alerts are checked by the company's human analysts before police are notified, flagged a student holding a clarinet as a possible gun. According to a police report reviewed by The Washington Post (as relayed by Ars Technica) and obtained by WKMG News 6, officers expected an armed person in camouflage holding a shouldered rifle; dispatch later told them the suspected rifle was likely a band instrument, and police found the student, who was dressed as a military film character for a themed dress-up day, among panicked students hiding in the band room. Officers questioned him and found no weapon in his backpack or instrument case (Oviedo Community News, February 2026). The student said he was unaware the way he held the clarinet could trigger the alert. ZeroEyes said it had not made an error, while also treating the alert as a true positive only until officers on the ground found no rifle and then classifying it as a false positive; the district said the system worked as intended; the principal asked parents to talk to their children about the dangers of pretending to have a weapon.
Contextual tracker case Low reported severity Involving minors Media Coverage
AI involvement reported · Causal attribution supported · 4 sources, 3 underlying accounts · Added 23/10/2026
30 Sept 2024Event location unknownBaidu AI smart answer
A practising lawyer in Nanjing, China, found in September 2024 that Baidu search's 'AI smart answer' (AI 智能回答) feature, asked how many years he had been sentenced to, answered that he had been sentenced to three years in prison for a criminal offence and showed his photograph in a lawyer's robe. According to the first-instance judgment as quoted in the press, the answer was produced on 30 September 2024. The lawyer says he complained repeatedly to Baidu by letter, hotline and email without the content being fully removed, and he sued. The Jiangbei New Area People's Court in Nanjing found that Baidu had combined text and his picture through AI into false, disparaging content aimed at him, that Baidu was at fault, and that this was defamation; it ordered a written apology and rejected his claims for lost income and emotional damages for lack of evidence. Baidu argued that AI hallucinations could not be foreseen. The Nanjing Intermediate People's Court dismissed Baidu's appeal on 12 March 2026. The lawyer applied for enforcement when no apology came, and in September 2026 said he had received Baidu's written apology; Baidu has applied for a retrial.
Contextual tracker case Medium reported severity
AI involvement supported · Causal attribution established · 3 sources, 2 underlying accounts · Added 22/10/2026
2020Event location unknownHireVue
City AM and Cosmetics Business reported in March 2022, relaying the BBC Three documentary Computer Says No and the Times, that three make-up artists at MAC, an Estée Lauder Companies brand, lost their jobs in a 2020 redundancy process after being told to reapply for their roles through a video interview run by HireVue and marked by an algorithm. City AM reports that they were made redundant partly as a result of the automated judgement. The women said no one could explain how they had been scored, and one said nobody watched the video. Estée Lauder settled their claims out of court; the amount was not disclosed. The company says staff were told in advance about the HireVue skills assessment, that the process was used in tandem with human decision-making, that facial recognition made up about 0.25% of the overall assessment and performance data and qualitative feedback most of it, and that any suggestion facial recognition played a decisive role is false. HireVue said its video-analysis component had been discontinued about two years earlier. How much the HireVue score weighed in each decision is not reported.
Contextual tracker case Medium reported severity
AI involvement reported · Causal attribution disputed · 2 sources, 1 underlying account · Added 22/10/2026