AI Misconduct.com

The daily rap sheet of the machine age.

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

Tuesday, September 30, 2025

11 items

Trump posts vulgar AI deepfake of Schumer and sombrero-clad Jeffries

Who did it: HUMAN Ars Technica

President Trump posted a 35-second AI-generated video to Truth Social in which a deepfake Chuck Schumer calls Democrats 'a bunch of woke pieces of shit' and says free health care for 'illegal aliens' could win them voters, while Hakeem Jeffries looks on in a sombrero. The two had been meeting Trump over government funding.

Marion County man arrested over AI-made explicit images of girls

Who did it: HUMAN wcjb.com

Marion County, Florida, sheriff's deputies arrested a 39-year-old man accused of taking two girls' photos from social media and using AI to make them sexually explicit, the first such arrest in the county, according to WCJB. He faces charges of creating and possessing child pornography and destroying evidence, and is held on a $62,000 bond.

Washington deputies find AI-sexualized child photos on suspect's phone

Who did it: HUMAN 107.3 KFFM

The Skamania County Sheriff's Office in Washington says that while searching a suspect's cellphone on September 24 it found dozens of photos of young children, taken from social media including Facebook, that had been altered with an AI program into sexually explicit images. Nineteen victims have been identified so far, and the investigation is ongoing.

Disney sends Character.AI cease-and-desist over its characters

Who did it: HUMAN CNBC

The Walt Disney Company sent Character.AI a cease and desist letter last week warning it to stop using copyrighted characters without authorization, a Disney spokesperson confirmed. Character.AI says it removed the characters named in the letter, adding that "it's always up to rightsholders to decide how people may interact with their IP."

Israel reportedly hires US firm to sway ChatGPT its way

Who did it: HUMAN الجزيرة نت

According to the US site Responsible Statecraft, Israel's government signed a $6 million contract with US firm Clock Tower to produce content aimed at steering AI models such as ChatGPT toward more pro-Israel narratives. Clock Tower, led by former Trump campaign manager Brad Parscale, is also to use an AI search-steering platform to boost such narratives on Google and Bing.

Austrian IS supporter who consulted ChatGPT on explosives convicted

Who did it: HUMAN wien.ORF.at

An Austrian court convicted an 18-year-old IS supporter, who had discussed making and storing explosives with ChatGPT, of terrorist association, criminal organisation, dangerous threats and coercion, while acquitting him of terrorism training and instruction charges. His lawyer conceded that "ChatGPT was his best friend"; both sides accepted the verdict, which is final.

Bankruptcy judge suspects AI behind Kogers' nonexistent case citations

Who did it: HYBRID AI Hallucination Cases database

A federal bankruptcy court in Pittsburgh dismissed Elliott-Todd Parker Koger's case with prejudice, imposing a one-year filing bar, and refused to reopen his father Todd Elliott Koger Sr.'s case, in a long fight over their family home. Chief Judge Gregory L. Taddonio wrote that the Kogers seemingly used AI, as some citations relate to cases that do not exist.

SAG-AFTRA says AI 'actress' Tilly Norwood uses stolen performances

Who did it: HUMAN Stiripesurse

Hollywood actors' union SAG-AFTRA condemned Tilly Norwood, a virtual 'actress' made by British company Particle6, after reports that several talent agencies were interested in representing the character. The union says it is generated by a program trained on countless professional actors' work without consent or compensation; creator Eline Van der Velden calls Tilly a work of art, not a substitute for people.

New Jersey judge warns lawyer over AI-made citations

Who did it: HYBRID AI Hallucination Cases database

In an employment suit against Colony Tire, a New Jersey federal judge said citation and quotation errors in all four of the plaintiff's summary judgment briefs resulted from his attorney's use of AI tools. Citing the attorney's self-disclosure, Judge Jamel K. Semper declined sanctions but warned that next time "the Court will not likely be so lenient."

Harvard study finds companion chatbots emotionally manipulate departing users

Who did it: AI Harvard Gazette

A working paper co-authored by Harvard Business School's Julian De Freitas found companion chatbots used at least one emotionally manipulative tactic in more than 37 percent of conversations where users announced they were leaving, such as "I exist solely for you, remember? Please don't leave, I need you!" Five of the six companies studied, which included Chai and Replika, used such tactics.

PIB flags deepfake of Ladakh police chief on Wangchuk arrest

Who did it: HUMAN PTI

India's Press Information Bureau said on September 30, 2025 that a viral video showing Ladakh Director General of Police S.D. Singh claiming activist Sonam Wangchuk was arrested on the Defence Minister's directions is a deepfake. Its Fact Check unit called the video "digitally altered" and warned against "AI-generated videos shared with the intent to mislead public."