AI Misconduct.com

The daily rap sheet of the machine age.

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

Wednesday, May 21, 2025

9 items

Publishers' trade group calls Google's AI Mode 'theft'

Who did it: HUMAN The Verge

Google's newly expanded AI Mode "just takes content by force and uses it with no return, the definition of theft," said Danielle Coffey of the News/Media Alliance, urging that DOJ remedies address it. An internal document from Google's antitrust trial showed the company decided against asking publishers' permission to include their work in AI features, Bloomberg reported.

Sun-Times summer reading list recommends books that do not exist

Who did it: HYBRID Chicago Sun-Times

The Chicago Sun-Times' May 18 'Heat Index' special section, licensed from Hearst's King Features, included a summer reading list created with an AI tool that recommended books that do not exist. King Features said a freelancer used AI without disclosing it and is ending the relationship; the paper pulled the section from its e-paper and won't charge subscribers.

Character.AI and Google must face teen suicide lawsuit, judge rules

Who did it: HYBRID Reuters

A federal judge ruled that Character.AI and Google must face a mother's lawsuit alleging the startup's chatbots caused her 14-year-old son's suicide. U.S. District Judge Anne Conway said the companies failed to articulate "why words strung together by an LLM" are speech. Character.AI says it will keep fighting; Google says the two firms are "entirely separate."

South Korean police detain teen over 500 deepfake nude images

Who did it: HUMAN 매일경제

Police in South Gyeongsang province have detained a teenage high school student suspected of pasting the faces of female celebrities and members of the public onto some 500 nude images and videos between last November and April, then sharing them in three Telegram chatrooms. Another 23 people, 15 of them teenagers, were booked for making or redistributing about 3,500 fake images and videos.

Israeli teen who allegedly asked ChatGPT about attacks indicted

Who did it: HUMAN כאן

Israeli prosecutors have indicted a 16-year-old for attempting to cause grievous harm as an act of terror, alleging he burst into his town's police station and swung a knife at a Border Police officer. According to prosecutors, he had consulted ChatGPT beforehand about possible attacks and methods. They are seeking his detention until the end of proceedings.

Deepfake Norman Swan peddles unproven supplements on Facebook

Who did it: HUMAN ABC News (Australia)

Deepfake videos and articles of ABC health journalist Norman Swan on Facebook and Instagram are promoting unproven supplements such as Glyco Balance and Keto Flow + ACV, the ABC reports. One fake article told diabetes patients to replace metformin with Glyco Balance, and one reader with diabetes stopped his prescribed medication. Meta says scammers operate "at industrial scale."

Surge AI sued for allegedly misclassifying data annotators

Who did it: HUMAN Los Angeles Times

A proposed class action filed Monday accuses San Francisco-based Surge AI, which hires "data annotators" to improve AI text responses for Meta and OpenAI, of deliberately classifying them as independent contractors to deny employee benefits. The complaint alleges unpaid training and near-impossible task time limits that docked pay. Surge AI did not respond to a request for comment.

Judge strikes anonymous filer's objections citing nonexistent case

Who did it: HYBRID E.D. Michigan

U.S. District Judge Judith E. Levy struck objections by an anonymous, self-represented non-party that cited "Navarro v. Sedgwick Claims Mgmt. Servs.," a case the court could not locate in any court and which bore "the hallmarks" of AI. The court warned all litigants in the case that citing nonexistent cases may bring sanctions and allowed a refiling, with checked citations, by June 3, 2025.

Windows 95 chime composer Brian Eno rebukes Microsoft's Israel AI deals

Who did it: HUMAN Pitchfork

Brian Eno, composer of the Windows 95 start-up chime, published an open letter, "Not in My Name," urging Microsoft to suspend services supporting violations of international law, after a May 15, 2025 company blog post acknowledged supplying Israel's Ministry of Defence with Azure cloud and AI services. Eno pledged his original chime fee to victims of the attacks on Gaza.