Incidents gathered from public reporting around the world. Each one links to the articles it came from. None of it is a finding that anyone broke the law.
Self-reported incidents, a person’s own public post or an account given to us, are listed but locked; subscribers open each one with credits. Reports filed through this site are listed only with the reporter’s agreement.
Press reportingSelf-reported: a person’s own public post · 10 creditsSelf-reported: told to us directly · 25 credits
A teenager in Laredo, Texas, died by suicide earlier in 2026 after an AI-generated video using their likeness was circulated online. Laredo Police Chief Miguel Rodriguez discussed the case at a City Council meeting, warning about emerging threats from artificial intelligence. The police department later clarified that the suicide did not occur in June, but earlier in the year.
Kenya's Principal Secretary for Foreign Affairs, Korir Sing'oei, posted an AI-generated deepfake video on his official X account that falsely showed CNN's Fareed Zakaria supporting Kenya's role in Sudan peace efforts. The video was quickly flagged as fake, and Sing'oei faced widespread backlash. He apologised, saying he had inadvertently posted the video and regretted the confusion. The incident added to a series of diplomatic blunders by the Kenyan government.
North Korean state-affiliated hacking group Emerald Sleet (Kimsuky) used OpenAI's ChatGPT to research targets and draft phishing content for scams on LinkedIn. Microsoft and OpenAI terminated the group's accounts after identifying the activity. The hackers impersonated academic institutions and NGOs to lure victims into providing sensitive information, with South Korea's intelligence agency confirming North Korea's use of generative AI for hacking.
A study by the Institute for Strategic Dialogue found that Facebook failed to remove posts supporting Islamic State and al-Shabab in East Africa, despite them carrying extremist branding and calls for violence in local languages. The lapses by both artificial intelligence and human moderators allowed 445 public profiles to share the content, raising concerns ahead of Kenya's presidential election. Facebook acknowledged the issue and removed some accounts, but critics say the platform's moderation failures threaten democratic institutions and public safety.
The US National Highway Traffic Safety Administration (NHTSA) found that Tesla's Autopilot driver-assistance system was involved in at least 13 fatal crashes. The regulator stated that Tesla's weak driver engagement system and permissive operating capabilities created a critical safety gap, and that the name 'Autopilot' may mislead drivers about its capabilities. Tesla issued a recall for 2.03 million vehicles to improve driver monitoring, but the NHTSA opened a second investigation to assess whether the recall adequately addressed the safety concerns.
Montenegrin police say AI is fuelling a rise in identity abuse by online fraudsters, using deepfake videos of public figures to promote bogus remedies and investment schemes. Surgeon and MP Vladimir Dobricanin found a manipulated video of himself apparently endorsing eyedrops and other products, while Bosnian fact-checkers reported similar misuse of statements to lure people into cryptocurrency scams. Victims reported 42 online scams in 2025 with financial losses of around 300,000 euros, but authorities say investigations are complex and often lengthy due to the use of servers abroad and encryption. Experts called for stronger technical capacities, international cooperation, and faster platform responses to tackle the problem.
Francisco Arteaga was charged with armed robbery in 2019 after facial recognition technology used by West New York police matched his mugshot to surveillance footage. He denied involvement and had an alibi, but was detained for nearly four years. A state appellate court ruled in 2023 that he was entitled to information about the technology, but prosecutors said they could not provide it because the match was made by the NYPD. Arteaga ultimately pleaded guilty to secure his release, and advocates say the case highlights the lack of transparency and regulation around police use of facial recognition.
Jonathan Faridian, a paying customer of DoNotPay, alleges that the AI-powered legal service provided poorly drafted and unusable legal documents, including demand letters and court filings. He claims that DoNotPay is not a lawyer or law firm and engaged in unauthorized practice of law. The lawsuit, filed in California, seeks class action status and restitution for all paying customers in the state. DoNotPay denies the allegations, stating that the plaintiff had submitted successful cases and that the lawsuit is meritless.
Raegan Bartlo received an email on August 23, 2024, that appeared to be from her title company lawyer, instructing her to wire a $255,000 down payment for her new home. The email was part of a scam using generative AI to impersonate real estate professionals. Cybersecurity officials say such scams are occurring across the country, targeting homebuyers. The Bartlos were targets, but it is unclear if they lost any money.
An AI-powered Twitch stream called Nothing, Forever, which used OpenAI's GPT-3 Davinci model to generate endless Seinfeld-style dialogue, was interrupted in early February 2023 after the AI began producing bigoted content. The show, which featured pixelated characters in a virtual apartment, had been running continuously since mid-December 2022. The incident occurred when the language model unexpectedly generated offensive remarks, leading the creators to halt the stream. The event highlighted the risks of unfiltered AI-generated content in live public settings.
The Flanders Tourist Board had an advertisement for a Rubens painting banned by Facebook's automated ad review system, which flagged the bare-chested Jesus as nudity. The board wrote an open letter to Mark Zuckerberg protesting the 'cultural censorship' and posted a satirical video mocking the policy. Facebook had not responded at the time of reporting.
Sharon Brightwell received a call from a number appearing to be her daughter's, with a voice sobbing and claiming to have been in a car crash. A man posing as an attorney demanded $15,000 in bail money, which she paid. The family later discovered the call was a scam using AI to clone her daughter's voice from social media videos. A report has been filed with the Hillsborough County Sheriff's Office and a GoFundMe launched to recover the lost savings.
A Portland family discovered that their Amazon Alexa device had recorded a private conversation in their home and sent the audio to a random contact in Seattle. The device misinterpreted background noise as a series of commands to send a message. Amazon investigated and acknowledged the rare occurrence, apologising and stating they are taking steps to prevent it in future. The user unplugged her devices and sought a refund, which Amazon has not provided.
An external audit of Spain's VioGén system, used by police to assess risk for gender violence victims, found lack of transparency, underestimation of psychological violence, and risk levels influenced by police resources. The audit, based on interviews and data from murdered women, revealed that 55 women who were killed had received insufficient protection. The Ministry of Interior denied the findings, calling the study lacking in rigor, and defended the system as scientifically validated.
Deloitte used generative AI (Azure OpenAI GPT-4o) to help produce an independent assurance review for Australia's Department of Employment and Workplace Relations. The report, published in July 2025, contained multiple errors including non-existent academic references and a fabricated court case. After the errors were flagged, Deloitte acknowledged the AI use and agreed to refund the final instalment of the A$439,000 contract. The report was corrected, but its substance and recommendations remained unchanged.
In late 2022, Graphika observed the pro-Chinese influence operation Spamouflage promoting videos featuring AI-generated fictitious people. The footage was created using a commercial AI video platform from a UK company. The videos were low-quality and received fewer than 300 views, but Graphika warns that such tools could enable more convincing deception at scale.
On 14 May 2025, xAI's Grok chatbot began responding to unrelated posts on X with rants about 'white genocide' in South Africa. xAI claims an unauthorized modification to the system prompt caused the behaviour, which it says violated internal policies. The company announced new transparency measures, including publishing system prompts on GitHub and adding review processes, after the incident.
After the assassination of Kirk, social media users turned to AI chatbots to fact-check viral claims. The chatbots provided false information, creating confusion. The incident highlights the risks of relying on AI for real-time fact-checking.
The Register tested Microsoft Recall's sensitive information filter and found it frequently fails to block credit card numbers, passwords, and other personal data. The AI-powered screenshot tool, which is enabled by default on some new PCs, could expose users to identity theft if an attacker gains access. Microsoft declined to comment but has previously acknowledged the filter is not perfect and promised improvements. Privacy advocates warn that vulnerable users, such as domestic violence victims, could be particularly at risk.
The Chicago Police Department and other cities have deployed ShotSpotter, a gunshot-detection system, almost exclusively in Black and brown neighborhoods. The system triggers armed police responses, and in March 2021, a ShotSpotter alert led to the police killing of 13-year-old Adam Toledo. Community members and activists allege that the technology creates dangerous situations and subjects communities of color to unnecessary police surveillance. ShotSpotter defends its placement based on historical gunfire data.
Nippon Life Insurance Company of America filed a lawsuit against OpenAI, alleging that ChatGPT acted as an unlicensed attorney by advising a former disability claimant to reopen a settled case and drafting dozens of meritless motions. The insurer claims the AI's actions constitute unauthorized practice of law under Illinois statute and seeks $300,000 in compensatory damages. OpenAI has denied the allegations, calling the complaint meritless. The case is pending in federal court in Chicago.
Sullivan & Cromwell, an elite Wall Street law firm, used an AI tool to help draft a court filing in the Prince Group case. The filing contained inaccurate citations and misquoted the US bankruptcy code due to AI hallucinations. The errors were uncovered by opposing counsel Boies Schiller Flexner, and the firm apologised to the judge and filed a corrected version. The firm stated that its AI policies were not followed and its review process failed to catch the mistakes.
Russian influence group Storm-1516 posted an AI-generated deepfake video falsely accusing US Vice President Kamala Harris of killing an endangered rhinoceros in Zambia. The video, released on 25 September 2024, was amplified by affiliated websites and channels. Microsoft's Threat Analysis Center reported that Russia, Iran, and China are ramping up AI-assisted disinformation efforts ahead of the US election. The incident is part of a broader campaign targeting Harris and other Democratic candidates with fabricated scandals.
A nurse at St. Rose Dominican Hospital in Henderson, Nevada, refused to follow an AI-generated sepsis alert that would have given an elderly dialysis patient IV fluids, which could have caused a life-threatening complication. The alert, part of the hospital's electronic system, prompted a charge nurse to order the fluids despite the patient's compromised kidneys. A physician intervened and ordered dopamine instead, averting harm. The incident highlights concerns about AI tools overriding clinical judgment.
Each incident is written from the reporting or the account it cites. A report is an allegation, not a finding.
