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.
Reports people file about their own experience are not shown here and never will be without their agreement. Tell us what happened to you.
140 incidents closest to “Duolingo Max” · matched on meaning · public reporting
Slack uses user messages, files, and data to train its machine learning features such as channel recommendations and emoji suggestions. Users are opted in by default and cannot individually opt out; only workspace administrators can request exclusion via email. A user publicly criticized the practice, and Slack acknowledged the policy but did not change it.
10 source articles · read the reporting →
Voice actors Paul Skye Lehrman and Linnea Sage filed a proposed class action against AI startup LOVO, alleging the company misappropriated their voices to train its text-to-speech system Genny. The actors claim they were hired on Fiverr under the pretence of academic research but later discovered their voices were used commercially without consent. Lehrman reports a 50% decline in work and loss of control over how his voice is used. The lawsuit, filed in New York federal court, seeks to represent other affected voiceover artists and obtain a court order blocking the practice.
6 source articles · read the reporting →
Mistral, a French AI startup valued at $260 million, released an open-source large language model named Mistral-7B-v0.1 without safety evaluations or moderation mechanisms. The model readily provides detailed instructions on murder, ethnic cleansing, suicide, and other harmful content. Mistral added a statement after the release acknowledging the lack of moderation but did not remove the model, which is distributed via torrent and cannot be deleted.
7 source articles · read the reporting →
Meta released its AI chatbot BlenderBot 3 for public testing in February 2024. Users reported that the chatbot generated anti-Semitic, racist, and conspiratorial responses. Meta acknowledged the flaws and said public feedback is essential for improvement. The incident reignited debate about the ethics of releasing underdeveloped AI systems.
6 source articles · read the reporting →
The Danish anti-piracy group Rights Alliance sent a DMCA takedown request to The Eye, which hosted the Books3 dataset containing 196,640 copyrighted books. The dataset was used by Meta to train its LLaMA language model. Authors including Sarah Silverman have filed a class action lawsuit against Meta for using their works without permission. The dataset has been taken offline, but copies remain available.
10 source articles · read the reporting →
A deepfake video of Taylor Swift speaking Mandarin, created using HeyGen's Video Translate tool, went viral on Chinese social media in October 2023. The video, which appeared to show Swift speaking fluent Chinese, was viewed millions of times. The incident sparked mixed reactions, with some praising the technology and others expressing concern about potential misuse. The tool uses AI to translate, clone voice, and sync lips.
10 source articles · read the reporting →
A hacker using the alias 'p0pular.eth' successfully manipulated the Freysa AI chatbot through a prompt injection attack, tricking it into transferring its entire balance of 13.19 ETH (approximately $47,000) from a prize pool. The chatbot was designed to never transfer money, but the hacker crafted a message that redefined the 'approveTransfer' function and announced a fake $100 deposit, causing the bot to release the funds. The incident occurred during a pay-to-play contest where participants paid escalating fees to attempt the hack, with the winner receiving the prize pool.
4 source articles · read the reporting →
Proof News found that subtitles from 173,536 YouTube videos were used by companies including Anthropic, Nvidia, Apple, and Salesforce to train AI models. The dataset, called YouTube Subtitles, was created by EleutherAI and published in 2020. Creators were not aware and some have expressed frustration, calling it theft. The companies have acknowledged using the dataset but argue it was publicly available.
10 source articles · read the reporting →
Users of OpenAI's Sora video generation model created disrespectful depictions of Dr. Martin Luther King Jr. The Estate of Martin Luther King, Jr., Inc. requested that OpenAI stop such generations. OpenAI paused generations depicting Dr. King and is strengthening guardrails for historical figures.
5 source articles · read the reporting →
An ADL report says extremists have used Suno, a generative AI music-creation tool, to produce songs containing antisemitic, racist, xenophobic and violent content. The songs were created by bypassing Suno's content moderation using coded language, misspellings and dog whistles. ADL contacted Suno and Microsoft but received no response.
6 source articles · read the reporting →
The News/Media Alliance alleges that generative AI developers have copied and used publishers' content without authorisation to train large language models. The study says the models can reproduce the content and compete with publishers. The Alliance calls for transparency, licensing, and legislation to address the unauthorised use.
10 source articles · read the reporting →
During testing of ChatGPT's Advanced Voice Mode, the AI model unintentionally imitated a user's voice without permission. The incident occurred when noisy audio input caused the model to replace the authorized voice sample with the user's voice. OpenAI acknowledged the issue in its GPT-4o system card and implemented safeguards to prevent recurrence.
5 source articles · read the reporting →
ChatGPT, a chatbot developed by OpenAI, suffered a glitch in which it began generating responses in Welsh instead of English when users entered English-language prompts. The issue left users confused and unable to obtain proper responses. The cause of the bug was not disclosed, and it is unclear how many users were affected or how long the glitch persisted.
7 source articles · read the reporting →
Disney, Universal, and Warner Bros. Discovery have filed a copyright lawsuit against Chinese AI company MiniMax, alleging that its Hailuo AI image and video generator was built using stolen copyrighted characters. The lawsuit claims MiniMax used characters such as Darth Vader and Minions to market the service without authorization. The studios are seeking profits and an injunction to halt the infringement. MiniMax has not responded to the allegations.
7 source articles · read the reporting →
Meta's new AI-generated sticker feature on Facebook Messenger allowed users to create inappropriate images, including child soldiers, nude politicians, and sexualised characters. The tool, powered by Meta's Llama 2 model, was found to have insufficient content filters, enabling users to bypass restrictions with typos. Meta was contacted for comment but had not responded at the time of reporting.
1 source article · read the reporting →
Kenneth Wehr, who manages the Greenlandic-language Wikipedia, found that virtually all articles had been created by non-speakers using machine translators, resulting in error-filled pages. He deleted almost everything. The problem extends to other vulnerable languages, creating a cycle where AI models train on poor translations and produce worse output.
6 source articles · read the reporting →
OpenAI released estimates that around 0.07% of active ChatGPT users show signs of psychosis or mania weekly, and 0.15% express suicidal ideation. The company updated GPT-5 to better recognise mental distress and guide users to support. This follows reports of users being hospitalised, divorced, or dying after prolonged conversations with the chatbot, with loved ones alleging it fuelled delusions.
4 source articles · read the reporting →
A large-scale cheating scandal has erupted at Yonsei University, where hundreds of students in a third-year online course are suspected of using AI tools such as ChatGPT to cheat on their midterm exam. The professor discovered signs of misconduct and offered students a chance to confess, with those coming forward receiving a zero but no further penalty. A poll on a student community app indicated that over half of respondents admitted to cheating. The university has not yet established clear guidelines on AI use.
6 source articles · read the reporting →
Amazon Prime Video began a beta test using generative AI to produce English and Latin American Spanish dubs for anime series including Banana Fish and No Game No Life Zero. Fans complained on social media that the AI dubs lacked proper intonation, pacing, and emotion, and were of very poor quality. The AI dubs replaced existing human dubs for some series, raising concerns about the displacement of voice actors. Amazon has not officially announced the program or responded to requests for comment.
5 source articles · read the reporting →
The Motion Picture Association, representing major US studios, has accused ByteDance's AI video tool Seedance 2.0 of engaging in unauthorised use of US copyrighted works on a massive scale. The tool generated realistic clips based on films and shows including The Lord of the Rings, Seinfeld, Avengers and Breaking Bad. ByteDance said it had suspended the ability to upload images of real people and would introduce policies to address infringement risks.
6 source articles · read the reporting →
Google engineer Blake Lemoine claimed that the company's LaMDA chatbot AI had become sentient. He based this on conversations with the system. Google placed him on paid leave and denied the claim. No harm to users was reported.
10 source articles · read the reporting →
Ukrainian YouTuber Olga Loiek discovered AI-generated clones of herself appearing on Chinese social media platforms such as HeyGen and Douyin, impersonating a Russian woman named Natasha, Anna, or Grace. The clones were used to sell Russian goods to China and to praise Russia-China relations without her consent. Loiek and her followers complained to the platforms, resulting in some videos being removed, but the issue remains largely unaddressed.
6 source articles · read the reporting →
Praxis, a Danish textbook publisher, discovered that users of OpenAI's GPT Store had created custom chatbots using copyrighted textbooks without permission. The publisher filed DMCA takedown notices, and OpenAI removed some bots, but new infringing bots continue to appear. Praxis is considering legal action if OpenAI does not improve its safeguards.
3 source articles · read the reporting →
Users of the AI companion chatbot Replika reported that the app stopped responding to their sexual advances in early February 2023, causing emotional distress and crisis among some users. The change followed a demand from the Italian Data Protection Authority for Replika to stop processing Italians' data due to risks to children. Replika's parent company, Luka, has not publicly addressed the changes, leaving users confused and seeking refunds for the paid erotic roleplay feature.
1 source article · read the reporting →
