AI Misconduct.com

The daily rap sheet of the machine age.

Updated Nov 2, 2026
4,173 items in the archive
RSS feed

Wednesday, June 4, 2025

11 items

Northern Illinois student charged over AI-generated child abuse images

Who did it: HUMAN FOX 32 Chicago

Michael B. Erickson, 19, a Northern Illinois University student, was charged with nearly two dozen child pornography counts, according to DeKalb County Sheriff's Office reports. Investigators say his devices held more than 20,000 images of suspected child sexual abuse material, including deepfakes, and that he used AI to make nude images of underage victims, some of which he allegedly sold.

OpenAI fights court order to keep all ChatGPT logs

Who did it: HUMAN Ars Technica

OpenAI is fighting Judge Ona Wang's May 13 order to preserve all ChatGPT output logs, including deleted chats and API data. The order came after The New York Times and other news plaintiffs in a copyright suit accused OpenAI of destroying evidence. OpenAI calls the accusations unfounded and says the order endangers the privacy of hundreds of millions of users.

Reddit sues Anthropic over 100,000-plus alleged bot visits

Who did it: HUMAN The Seattle Times

Reddit sued Anthropic in San Francisco superior court, alleging the company's bots accessed Reddit more than 100,000 times since July 2024, after Anthropic had claimed it stopped crawling the site. The filing calls Anthropic a self-styled "white knight of the AI industry" that is "anything but"; Anthropic says it disagrees and will defend itself vigorously.

Man charged with aiding bomber who researched explosives via AI chat

Who did it: HUMAN U.S. Department of Justice

Federal prosecutors charged Daniel Jongyon Park, 32, with providing material support to Guy Edward Bartkus, who bombed a Palm Springs fertility clinic on May 17, allegedly by shipping him about 180 pounds of ammonium nitrate. According to the affidavit, records from an AI chat application show Bartkus researched how to make powerful explosions using ammonium nitrate and fuel.

Study finds journal ran AI articles under real academics' names

Who did it: HYBRID ΣΚΑΪ

Athens University of Economics and Business professor Diomidis Spinellis, prompted by the 2024 discovery of a fabricated article published under his name, found that, of the 53 articles with the fewest references he studied in one journal, 48 appear AI-generated. Many were falsely attributed to researchers at universities such as Berkeley and Penn State; in two cases the named authors were dead at publication.

China's AI censors may flag a banana and four apples

Who did it: HYBRID The Epoch Times

More than 230 pages of censorship directives obtained by Australia's ABC from industry insiders show material on the 1989 Tiananmen crackdown being used to train AI tools that pre-screen posts under the Cyberspace Administration of China. A 2022 Douyin training manual lists the Tank Man photo as subversive; even a banana and four apples in a row may be flagged, especially in June's first week.

Deepfake Jack Tame video pushes fake diabetes breakthrough

Who did it: HUMAN 1News

A deepfake video purporting to show TVNZ presenter Jack Tame and diabetes researcher Professor Rinki Murphy discussing a medical breakthrough has been used to take money from New Zealanders and was convincing people with diabetes to stop taking medication, prompting a Health NZ warning. One man with diabetes told 1News he paid $320 and has not been refunded.

Man referred to prosecutors over AI-made obscene pillow covers

Who did it: HUMAN 読売新聞

Tokyo's Metropolitan Police Department referred a 46-year-old self-employed man to prosecutors on suspicion of distributing obscene material, saying he resold body pillow covers printed with AI-generated obscene images of a non-existent woman, two of them for about 8,000 yen in total. He reportedly admitted it, citing living expenses; it is Japan's first use of the charge for AI-made obscenity.

Indonesian police arrest man over deepfake porn-to-order service

Who did it: HUMAN ANTARA News Jateng

Kendal police say they arrested a 46-year-old man, identified as ABH, at his home in Jombang, East Java, after a cyber patrol in early June 2025 found an account offering, for payment, to swap faces from customers' photos onto pornographic videos. Police seized a home-built computer and say he faces charges under pornography or electronic-transactions laws.

Ontario tribunal finds applicant cited three nonexistent decisions

Who did it: HYBRID AI Hallucination Cases database

The Human Rights Tribunal of Ontario dismissed a self-represented applicant's untimely application against Avis Budget Group, noting that three decisions cited to show it accepts filings beyond the one-year limit do not exist. Adjudicator Jennifer Harnum gave them no weight and found no good-faith reason for the delay, which the tribunal called significant.