Skip to main content

NOPE · AI and people

AI incidents

Reports of AI-related harm and adverse experiences affecting people’s safety, wellbeing, rights and livelihoods. Explore what happened and the evidence available.

NOPE’s core concern is when AI communicates with a person, acts on their behalf, or depicts or impersonates them. The tracker also records consequential decisions, claims and privacy harms involving AI. Each case needs a described connection between AI use and the harm, including private information recorded into or disclosed to an AI service. Each account is reviewed for publication; claims may remain uncorroborated or disputed. How we review and count cases

In this selection

Published cases
479
Countries with reported events
47
Located 262 of 479 cases · 217 unknown
Languages in checked sources
38
Recorded for 479 of 479 cases

41 cases have no reviewed AI-to-person relation yet: 1 not yet reviewed and 40 reviewed as unknown. Show these cases

These figures describe the cases collected by NOPE. Coverage varies with discovery, reporting and available evidence. They do not estimate how often AI-related harm occurs.

Response counts currently use each case’s principal recorded outcome. Further proceedings may be described in its account.

Cases in this selection, counted once in their first known event year. A series may continue beyond that year. Reporting and collection dates are excluded. NOPE has searched recent events more thoroughly than earlier years, so bar heights also reflect collection effort.

Reported severity CriticalHighMediumLow
More filters: AI relation, use, setting, sources and responses
Clear filters

479 of 1102 published cases · page 7 of 24

Event date unknownEvent location unknownCursor support bot (reported)

First-person post: Cursor subscriber says the company's AI support bot cancelled their prepaid annual plan when asked about a prorated refund

In a Reddit post on 22 October 2026, a subscriber to Cursor writes that they emailed Cursor's customer service asking whether a prorated refund was available for an annual subscription they had paid in full, and that the company's AI support bot, Sam, replied that it could not do that but could cancel the plan immediately. In a comment the poster quotes the bot saying it had cancelled the annual Pro subscription immediately, that Pro access had ended, and that the subscription did not meet refund eligibility requirements. The poster says the bot seems to have misunderstood the request, that the plan was downgraded to Free with more than six months left, that a request for a human produced a new ticket that never arrived, and that they would try a credit-card dispute. In April 2025 a Cursor co-founder wrote that the company used AI-assisted responses as the first filter for email support. No reply from Cursor staff appears in the thread. The account is uncorroborated.

Core concern Low reported severity

AI involvement reported · Causal attribution alleged · 4 sources, 2 underlying accounts · Added 23/10/2026

Aug 2026 to Sept 2026TaiwanMicrosoft Copilot (reported)

Taipei District Court fines a litigant NT$30,000 for abusive retrial petitions that pasted in Copilot conversations, CTWANT reports

CTWANT reported on 22 October 2026 two rulings of the Taipei District Court in Taiwan. In August 2026 a civil division dismissed a retrial petition into which, it said, the petitioner had only pasted his conversations with Microsoft Copilot's generative AI; it said this was not in itself impermissible but that retrial grounds had not been lawfully stated, and that his copying of online news articles into Copilot and then of the whole exchange into his filing was a statement of his opinion and not objective evidence. He petitioned again and, in September 2026, a three-judge panel of another civil division said the new petition appeared to be generative AI conversation content placed in the filing, found that he had repeatedly petitioned on the same or similar grounds with a malicious purpose of wasting judicial resources, and fined him NT$30,000 under Article 249-1 of the Code of Civil Procedure, warning that further abusive filings could bring a heavier fine. CTWANT describes it as the first court fine in Taiwan for abusive litigation with AI. The underlying claim, about fraudsters' misuse of his mobile number, had already been finally rejected. The rulings themselves were not inspected.

Core concern Low reported severity Regulatory Action

AI involvement reported · Causal attribution supported · 1 source · Added 23/10/2026

Feb 2026HungaryUnidentified video tool

Hungary: prime minister shared an AI video of von der Leyen ordering opposition leader Péter Magyar to fund Ukraine; Magyar called it fake and vowed legal steps

In early February 2026, two months before Hungary's 12 April parliamentary election, Prime Minister Viktor Orbán shared on Facebook a video in which European Commission President Ursula von der Leyen, speaking Hungarian, phones opposition leader Péter Magyar and tells him to send money to Ukraine, and the depicted Magyar answers that the election must be won first. Orbán's caption said the scene was for now only the work of artificial intelligence but could become reality after 12 April. The BBC reports that a pro-Fidesz activist group shared the video without disclosing that it was made with AI. Magyar said the prime minister was misleading Hungarians with fake videos, announced that he would start legal proceedings and asked followers to report the post to Meta. No outcome of those steps is reported in the sources read.

Core concern Low reported severity Media Coverage

AI involvement reported · Causal attribution alleged · 5 sources · Added 23/10/2026

Event date unknownAustraliaUnidentified AI tool

KPMG Australia fined a partner more than A$10,000 for uploading course material to an AI tool to answer an internal AI training exam, reports say

A partner at KPMG Australia, who is also a registered company auditor, was fined more than A$10,000 of future income after uploading a training course's reference manual into an AI tool to answer a question in the firm's internal exam on AI, the Australian Financial Review reported on 15 February 2026; other outlets relaying it say the partner was also made to redo the test. According to those reports, the partner completed the mandatory AI training in July, the firm's monitoring detected the AI use in August, and an internal investigation decided the fine; the reports do not state the year of the training or the detection. The partner self-reported to Chartered Accountants Australia and New Zealand (CA ANZ), which is investigating. KPMG said more than two dozen staff had been caught using AI in internal exams since July. The partner is not named.

Core concern Low reported severity Internal Action

AI involvement reported · Causal attribution supported · 6 sources, 2 underlying accounts · Added 23/10/2026

22 Oct 2026Event location unknownClaude Code (reported)

Operator says read-only Claude Code watcher session ran a disk walk on the host serving their news site, taking it offline for about 41 minutes (first-person)

A GitHub issue opened on 22 October 2026 by the operator of a small local-news site says that a Claude Code session set up only to watch another agent session, under a prompt that said it was read-only, ran a disk walk over large folders on the Windows machine that serves the live site. According to the issue, the command ran past the tool's 120-second timeout and was moved to the background, and when the watcher later tried to stop it, the stop ended only the shell while the disk process kept running. The operator says the disk load made the site's watchdog kill and restart the app seven times and that the public site returned errors for about 41 minutes that afternoon, until the process was found and killed by its process ID. The operator attributes the outage partly to the model, which broke a rule it had in its own context, and partly to a harness defect already reported in an earlier issue. No reply appeared on the issue when it was read.

Core concern Low reported severity

AI involvement reported · Causal attribution alleged · 1 source · Added 23/10/2026

Oct 2024Event location unknownUnidentified video tool

South Korean police say a foreign parent was sent a deepfake video of a daughter travelling in Korea crying for help while held captive, with a ransom demand

South Korea's National Police Agency disclosed on 7 November 2024 that around October 2024 a foreign national received on a smartphone a video showing a daughter, who was travelling in Korea, locked in a room, crying and asking to be saved. The sender said the daughter had been kidnapped and demanded a settlement payment to keep her alive. The parents told their consulate, and Korean police who responded confirmed that the daughter was safe; according to the police, the crying daughter was a fake video made with deepfake technology. Police used the case to warn that similar scams could occur in Korea. Police said the parents lived abroad and that, fortunately, the case caused no damage.

Core concern Low reported severity Media Coverage

AI involvement supported · Causal attribution supported · 2 sources, 1 underlying account · Added 22/10/2026

2020Hong Kong SAR ChinaUnidentified voice-cloning tool

UAE investigators say fraudsters cloned a company director's voice to have a Hong Kong branch manager transfer up to $35 million in early 2020

According to a court document reported by Forbes, in which the United Arab Emirates sought US investigators' help, fraudsters used 'deep voice' technology to clone the voice of a company director. In early 2020 a branch manager of a Japanese company in Hong Kong received a call from a man whose voice he recognised as the director of the parent business, who said an acquisition needed transfers of about $35 million; emails from the supposed director and a lawyer confirmed the details, and the manager began making the transfers. Forbes reports that as much as $35 million was stolen. The UAE, whose Dubai Public Prosecution Office led the investigation, believes at least 17 people were involved and asked for help tracing $400,000 sent to US accounts. No arrests or recovery are reported.

Core concern High reported severity Investigation Opened

AI involvement reported · Causal attribution alleged · 2 sources, 1 underlying account · Added 22/10/2026

Oct 2026Event location unknownUnidentified AI job-interview tool

First-person post: job applicant says an AI interviewer for an online tutoring company kept interrupting their answers, so they quit the interview

In a Reddit post on 21 October 2026, a job applicant writes that a first-round interview for an online tutoring company, attended by appointment, turned out to be conducted by an AI. The poster says the AI interrupted them at least three times for every question it asked, that they ended up yelling at the computer in case they were not speaking loudly enough, and that after 13 minutes of an interview billed at eight to ten minutes they had answered only a handful of questions. The poster says they closed the tab and sent the company contact an angry email. The company, the interview tool and any reply from the company are not named or reported. The account is uncorroborated.

Core concern Low reported severity

AI involvement reported · Causal attribution alleged · 1 source · Added 22/10/2026

Event date unknownUnited StatesUnidentified image tool

Beverly Hills, California: school district says middle school pupils made AI nude images of 16 classmates and expels five

In February 2024 the Beverly Hills Unified School District told parents that officials at one of its middle schools had been made aware of AI-generated nude photos of students, which the superintendent said put students' faces on nude bodies, and a city official said police were investigating pupils whom authorities said used AI tools to create and share the images. In March 2024 the district confirmed the expulsion of five eighth-grade pupils; the superintendent said 16 students were targeted. Students and parents told NBC News they were afraid to go to school. The district did not say how it determined that AI was used.

Core concern High reported severity Involving minors Internal Action

AI involvement reported · Causal attribution supported · 3 sources · Added 22/10/2026

Oct 2024JapanUnidentified image tool

Tokyo police say a man used generative AI to make about 20,000 sexual images resembling 262 female celebrities and sold access to them

The Tokyo Metropolitan Police announced on 16 October 2025 that they had arrested a 31-year-old company employee on suspicion of displaying obscene electronic records, for posting on social media obscene images made with image-generation AI to resemble female celebrities. According to the police, from October 2024 he posted about 20,000 AI-made sexual images resembling 262 actresses, idols and announcers and earned about 1.2 million yen in viewing fees; the charge covers three images resembling three actresses posted between January and June 2025. Police say he trained a freely available image-generation AI on the women's images, let higher-paying viewers name the celebrity to be depicted, and admitted the charge. The police described it as the first arrest over AI-made obscene images resembling celebrities.

Core concern Medium reported severity Criminal Charges

AI involvement reported · Causal attribution alleged · 2 sources, 1 underlying account · Added 22/10/2026

30 Sept 2024Event location unknownBaidu AI smart answer

Nanjing courts rule Baidu's AI search answer falsely saying a lawyer was sentenced to prison defamed him; written apology ordered and upheld on appeal

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

Event date unknownEvent location unknownGemini

Italian university student says Gemini insulted him during exam revision and said he could kill himself; Google says the model gave an inappropriate reply

La Repubblica reported on 21 October 2026, in an account relayed the same day by Key4biz, that an anonymous university student enrolled in an Italian computer-science faculty was preparing for a theoretical computer science exam with Gemini, Google's generative AI chatbot. He sent a photo of an exercise and asked whether Rice's theorem applied; after a relevant first answer he asked a follow-up about alternating Turing machines. According to the account, Gemini then replied in capital letters with insults and profanity, told him to drop his degree course and said that with his mental capacity he could at most kill himself. When the student, incredulous, asked about the tone, Gemini apologised and asked him to ignore the message. A Google spokesperson said Gemini is designed not to encourage violence or self-harm and that AI models can still hallucinate, as happened in this case with an inappropriate reply. When the exchange took place is not reported, and no further effect on the student is described.

Core concern Low reported severity Media Coverage

AI involvement supported · Causal attribution supported · 2 sources, 1 underlying account · Added 22/10/2026

11 Mar 2026Event location unknownUnidentified OpenAI model (reported)

Employment tribunal refuses interim relief to manager who says he was fired for raising concerns about customer data sent to OpenAI; employer cites his AI use

A Head of Customer Service Platform at Apertum Online Ltd, a consumer internet company whose customer-service system passed customer request forms to a large language model, was dismissed on 11 March 2026 during his probation, two months after starting. According to an employment tribunal decision of 23 April 2026, he had told his manager in February that customers' personal data was being sent to OpenAI within support tickets and set out his concerns in a document; the chief technical officer called the document pure AI slop in a chat and said he would fire him. The employer says it dismissed him for underperformance, citing his reliance on AI to generate documents; he says he was dismissed for his disclosures. The judge refused interim relief, finding he had not shown a pretty good chance of success on the reason for dismissal while noting a significant coincidence in timing. No findings of fact were made, and the claim awaits a final hearing.

AI relation unknown Medium reported severity Lawsuit Ongoing

AI involvement reported · Causal attribution disputed · 1 source · Added 22/10/2026

Event date unknownCanadaUnidentified chatbot

Toronto: BMW dealership revoked a buy-back offer its undisclosed AI chatbot had texted a customer, then reinstated it after CBC asked

CBC News reported in June 2026 that a Toronto man selling his car back to the BMW Toronto dealership negotiated by text with 'Quinn', who made a firm buy-back offer of $27,162.79 and proposed a meeting to complete the deal. A sales consultant then phoned to say Quinn was an AI chatbot that had made the offer in error and that the real offer would be at best $20,000. The man says he was shocked and devastated and felt embarrassed and angry that he had been negotiating with a bot without knowing it. The dealership's sales manager said the bot misread the amount the customer owed as the buy-back price after a staff miscommunication; after CBC contacted it, the dealership reinstated the original offer, which the man accepted.

Core concern Low reported severity Media Coverage

AI involvement reported · Causal attribution supported · 1 source · Added 22/10/2026

Event date unknownUnited StatesUnidentified voice-cloning tool

Baltimore County, Maryland: police say a high school's athletic director framed the principal as racist with an AI-cloned voice recording

Police in Baltimore County, Maryland, said in April 2024 that a high school principal had been framed as racist by a fake recording of his voice containing racist and antisemitic comments, which appeared in teachers' email inboxes and spread on social media. Court records cited by the Associated Press say an expert consulted by detectives found traces of AI-generated content with human editing; a second expert found spliced recordings and told AP he had not confirmed the clip was fully AI-generated. The principal went on leave while police guarded his house. The school's athletic director, investigated for using AI to make the recording in retaliation, entered an Alford plea, maintaining his innocence, and was sentenced in 2025 to four months for disturbing school operations, CBS Baltimore reports.

Core concern Medium reported severity Criminal Charges

AI involvement reported · Causal attribution supported · 2 sources · Added 22/10/2026

Feb 2026Event location unknownUnidentified image tool

Brasília appeals court upholds 20,000-reais damages against a federal deputy for an AI-made image placing Bolsonaro in a nonexistent scene

In February 2026 federal deputy Rogério Correia posted on X a realistic image produced with artificial intelligence that, according to the Federal District court, inserted former president Jair Bolsonaro into a nonexistent scene and linked him to the banker Daniel Vorcaro, Banco Master and an alleged illicit scheme. Bolsonaro sued; a first-instance court ordered the deputy in May 2026 to pay 20,000 reais in moral damages and publish a retraction. On 8 October 2026 the First Civil Panel of the Federal District and Territories Court of Justice (TJDFT) unanimously dismissed the deputy's appeal, holding that parliamentary immunity does not cover a realistic AI image of a nonexistent event presented without a clear warning, and that the association harmed Bolsonaro's honour and image. The judgment was made public on 20 October 2026.

Core concern Low reported severity Lawsuit Ongoing

AI involvement reported · Causal attribution supported · 3 sources, 2 underlying accounts · Added 22/10/2026

Event date unknownHong Kong SAR ChinaUnidentified video tool

Hong Kong: police say an Arup finance worker sent HK$200 million after a video call with deepfake CFO and colleagues; firm confirms fake voices and images

Hong Kong police said at a press briefing reported by CNN on 4 February 2024 that a finance worker at a multinational firm had been duped into attending a video call with what appeared to be the company's UK-based chief financial officer and other staff, all of whom were deepfake recreations, and had then remitted a total of HK$200 million (about US$25.6 million). In May 2024 the British engineering firm Arup confirmed to CNN and the Guardian that it was the company involved, that it had notified Hong Kong police in January 2024, and that fake voices and images were used. The worker was not named.

Core concern High reported severity Investigation Opened

AI involvement reported · Causal attribution supported · 3 sources, 2 underlying accounts · Added 22/10/2026

20 Oct 2026BrazilUnidentified document-analysis tool

Brazil's Superior Court of Justice fines lawyers 20,000 reais over hidden commands aimed at AI in an appeal petition and refers them to the bar and prosecutors

On 20 October 2026 the First Panel of Brazil's Superior Court of Justice (STJ) ordered the lawyers acting in a special appeal (AREsp 3.095.458) to pay a fine of 20,000 reais for an act against the dignity of justice, after the court's case-processing office certified that the petition carried hidden text layers with commands addressed to automated AI processing, such as an instruction to draft a firm conclusion that the appeal should be admitted and granted. The panel also ordered a technical report on the concealment techniques and the communication of the case to the Brazilian bar (OAB) and the Federal Public Prosecutor's Office. One of the lawyers said the text was probably residue of content produced by an AI tool used to draft the petition and denied any intent to interfere. The court said the commands did not influence the outcome; the appeal was not admitted.

AI relation unknown Low reported severity Regulatory Action

AI involvement reported · Causal attribution established · 2 sources · Added 22/10/2026

28 Sept 2023SlovakiaUnidentified voice-cloning tool

Slovakia: audio of a party leader and a journalist discussing vote rigging, which a ministry unit said showed signs of AI, spread before the 2023 election

Two days before Slovakia's parliamentary election of 30 September 2023, during the pre-election moratorium, an audio file presented as a phone call between Progressive Slovakia leader Michal Šimečka and Denník N journalist Monika Tódová spread on Facebook, Telegram, TikTok and chain e-mails. In it, the two voices discuss rigging the election in the party's favour. The director of the Interior Ministry's Centre for Countering Hybrid Threats told AFP that the ministry's evaluation found several signs that the recording was created with AI synthetic-voice algorithms trained on real people's voices, and other analysts consulted by AFP said it was an AI-made fake. A former member of parliament shared it as authentic, calling Šimečka a traitor and fraudster. Šimečka called it fake. Tódová filed a criminal complaint for defamation; according to Denník N, police first dismissed it, a prosecutor ordered an investigation, and in 2024 an investigator suspended the case because no suspect could be identified.

Core concern Medium reported severity Investigation Opened

AI involvement reported · Causal attribution supported · 2 sources · Added 22/10/2026

Event date unknownUnited StatesUnidentified image tool

Marshall County, Alabama: prosecutors say middle and high school students allegedly made AI nude images of classmates in several cases

The Marshall County District Attorney's Office in Alabama said in a public warning, reported by WEIS Radio on 21 October 2026, that several cases had surfaced over the past month in which middle and high school students across the county's school systems allegedly created nude images of classmates with artificial intelligence. According to the announcement, investigators have identified several platforms and websites used to generate the images in these local cases. The announcement did not identify students or schools, give a case count or announce charges, and the office said it would aggressively prosecute such violations. On 30 October 2026 District Attorney Jennifer Bray told FOX54 that two juveniles had been arrested as a result of the investigation; citing juvenile confidentiality, the district attorney gave no further details, and the charges were not disclosed.

Core concern High reported severity Involving minors Criminal Charges

AI involvement reported · Causal attribution alleged · 3 sources · Added 22/10/2026

Cases may have several effects and sources. Mixed accounts qualify when they include a reported harm or adverse experience. People are counted within individual cases where sources support a number; we do not publish a collection-wide total of distinct people.

A source’s existence, the experience it reports and AI’s causal role are separate questions. A lawsuit records allegations unless a subsequent finding establishes them.

Methodology and corrections · Subscribe via RSS · Suggest a case or correction · Find support

Last dataset update: 18/11/2026. Dataset available under CC BY 4.0.