AI Misconduct.com

The daily rap sheet of the machine age.

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

Wednesday, January 28, 2026

10 items

Illinois man charged in Wisconsin over AI-generated child pornography

Who did it: HUMAN WISN

Prosecutors in Waukesha County, Wisconsin, have charged a 20-year-old Illinois man with three counts of possession of virtual child pornography, alleging he used AI to manipulate Instagram photos of a teenage girl he knew to depict her naked. The case tests a state law criminalizing AI-generated child pornography; each count carries a minimum of three years in prison.

Fake vet call used AI-made puppy photo, Florida couple says

Who did it: HUMAN FOX 13 Tampa Bay

A Florida couple says a caller posing as a police sergeant claimed their runaway puppy had been hit by a car, sent a photo of the dog on an operating table they believe was AI-generated, and requested $1,900 for surgery via Zelle. The puppy turned up unharmed in their yard the next morning; their bank flagged the payment as fraud.

Music publishers sue Anthropic again, alleging pirated lyrics

Who did it: HUMAN Reuters

Universal Music Group, Concord and ABKCO sued Anthropic in California federal court on January 28, alleging it pirated more than 700 of their works, including lyrics and sheet music to "Sweet Caroline" and "Wild Horses", to train Claude. The publishers say they are suing over more than 20,000 songs, with potential statutory damages above $3 billion; Anthropic did not immediately comment.

Lawmaker says Nvidia helped DeepSeek train models China's military used

Who did it: HUMAN The Business Times

In a letter to Commerce Secretary Howard Lutnick, House China committee chair John Moolenaar says Nvidia records show its staff helped DeepSeek achieve major training efficiency gains in 2024 on models later used by China's military, though he notes there was then no public sign of such use. Nvidia says it makes no sense for China's military to depend on American technology.

MS NOW swaps thumbnail after airing AI-enhanced Alex Pretti image

Who did it: HUMAN InfoWars

MS NOW used an AI-enhanced image of Alex Pretti in a segment and as a YouTube thumbnail, later swapping the thumbnail with an editor's note saying the previous one was AI-enhanced. Megyn Kelly said the edit tanned him, whitened his teeth and shortened his nose; Mediaite's sources say a producer pulled the image off the internet unaware it was enhanced.

Reno officer concedes casino facial-recognition arrest was wrongful

Who did it: HYBRID Casino.org

In a January 22 deposition, Reno officer Richard Jager said his 2023 arrest of a truck driver at the Peppermill, after the casino's AI facial recognition misidentified him as a man previously trespassed from the property, "never should have happened". The driver, who is suing Jager and settled with the casino, spent 11 hours in custody; his lawyer seeks to add Reno as a defendant.

Phoenix facial-recognition cold-case murder charge quietly dismissed

Who did it: HYBRID ABC15 Arizona in Phoenix (KNXV)

Phoenix police arrested a Mexican man in a 1998 cold-case murder after investigators ran the suspect's old photo through facial recognition and picked his 2011 photo from 250 possible matches; the charge was dismissed on August 5, 2025, ABC15 reports. A judge ruled the detective gave misleading grand jury testimony, which omitted the use of facial recognition. Police say the case remains open.

Halifax man, 69, charged over AI-generated child abuse images

Who did it: HUMAN CTVNews

Halifax Regional Police say a 69-year-old man arrested Tuesday allegedly used artificial intelligence software to create and share child sexual abuse material, after investigators received information in September 2025. Police say evidence so far does not suggest real children were used or harmed. Michael Burke Leblanc appeared in Halifax provincial court on Wednesday to face charges.

Maryland federal judge orders lawyer to explain apparently fake citations

Who did it: HUMAN AI Hallucination Cases database

In a January 28 opinion in a woman's suit against Baltimore County police officers, US District Judge Brendan Hurson said perhaps as many as half of her lawyer's case citations appear inaccurate or non-existent, including a quoted, purportedly Fourth Circuit case whose citation most closely corresponds to a Ninth Circuit decision, and quotes absent from cited opinions. Counsel must respond within 21 days.

Chinese court finds developer not liable for chatbot's invented campus

Who did it: AI #SixthTone

The Hangzhou Internet Court dismissed a user's claim for nearly 10,000 yuan against an unnamed developer whose AI invented a nonexistent campus of a real university, insisted it existed, then promised him 100,000 yuan in compensation and suggested suing at that very court. The court ruled the AI cannot make legally binding expressions of intent and found no actual harm.