AI Misconduct.com

The daily rap sheet of the machine age.

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

Wednesday, October 8, 2025

8 items

Palisades fire suspect had ChatGPT draw burning city, prosecutors say

Who did it: HUMAN The San Francisco Standard

Federal prosecutors say Jonathan Rinderknecht, 29, charged with maliciously starting the Palisades fire on New Year's Day, used ChatGPT in July 2024 to generate an image of a city engulfed in flames and, after the fire started, asked whether you are at fault if a fire is lit because of your cigarettes. The blaze killed 12 people and burned 6,000 buildings.

IIIT Naya Raipur suspends student over AI-made obscene images

Who did it: HUMAN ETV Bharat News

A student at IIIT Naya Raipur is accused of using AI tools to turn social media photos of about 36 female students into obscene images. Police say over 1,000 fake photos and videos were recovered from his devices. The institute suspended him and opened an internal probe, which victims allege is an attempt to keep police out of it.

Brazilian police raid call centre selling AI-written prayers, arrest 35

Who did it: HUMAN O Globo

Police raided a call centre in Nilópolis, in the Baixada Fluminense, that they say sold believers in Brazil and abroad AI-generated prayers tailored to their troubles, for an average of R$50 each. Thirty-five people, including a pastor, were arrested and released on bail; police estimate the group took in R$3 million a year.

ACU relied on AI detector to accuse students of AI cheating

Who did it: HYBRID ABC News (Australia)

Australian Catholic University recorded nearly 6,000 alleged academic misconduct cases in 2024, about 90 per cent tied to AI, according to internal documents seen by the ABC, which also indicate ACU regularly relied on Turnitin's AI detector alone. Wrongly accused students say they waited months to be cleared. ACU calls the figures substantially overstated and dropped the tool in March.

Ukraine says Russian network pollutes AI models to target US

Who did it: HYBRID Укрінформ

Ukraine's Centre for Countering Disinformation, citing NewsGuard, says Russia has stepped up information attacks on the US using AI, through the Pravda network of more than 150 sites in 49 countries. The centre says the network spread over 3.6 million pieces of Kremlin-narrative content in a year and deliberately pollutes search engines and AI models so AI algorithms repeat Kremlin messaging.

Alicante court flags phantom Supreme Court quotes to bar association

Who did it: HYBRID AI Hallucination Cases database

An appeal brief before the Alicante Provincial Court attributed quoted passages to Supreme Court ruling STS 450/2025 that do not appear in it. Calling the brief the product of either AI hallucinations or a will to deceive, the court on 8 October 2025 ordered its ruling sent to the Alicante Bar Association for possible discipline, while partly upholding the appeal on other grounds.

Western Cape parents report AI-faked nudes of schoolgirls to police

Who did it: HUMAN IOL

Parents of learners at a high school on the Western Cape's West Coast have opened police cases after girls' faces were allegedly put onto nude pictures using AI and circulated at the school, the Western Cape Education Department confirmed. It said the school's own investigation could not substantiate charges against any particular learner.

Which? finds deepfake 'Quantum AI' scam videos on YouTube

Who did it: HUMAN Which?

Which? found eight YouTube videos, with over 17,000 views, promoting the 'Quantum AI' investment platform, which it says runs scams using deepfake videos of celebrities; YouTube removed all eight. A Reddit user reported an ad impersonating Prime Minister Sir Keir Starmer and Sky News, and the FCA warned earlier this year that Quantum AI may lack its permission.