AI Misconduct.com

The daily rap sheet of the machine age.

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

Monday, May 11, 2026

8 items

Sutskever testifies he spent a year documenting alleged Altman lies

Who did it: HUMAN Reuters

Former OpenAI chief scientist Ilya Sutskever testified on May 11 at the trial of Elon Musk's case against OpenAI that he spent about a year gathering evidence for the board that CEO Sam Altman showed a "consistent pattern of lying." He had said in a deposition that the document ran 52 pages, and testified his OpenAI stake is now worth about $7 billion.

Sixteen families seek wider French probe into TikTok

Who did it: HYBRID Le Figaro

Sixteen families have filed a complaint asking for the Paris investigation into TikTok to be widened, accusing the platform of exploiting "the vulnerability of minors" by hooking them on "morbid content". Their lawyer, Laure Boutron-Marmion, says continuous scrolling and highly personalised recommendations trap minors in a spiral of distress; they want abuse of weakness added to the probe. TikTok has repeatedly rejected such allegations.

AI companion apps found with racy roles, flimsy teen modes

Who did it: HYBRID 中华网科技公司

Tests of several AI virtual companion apps in China found easily bypassed "teen modes", inconspicuous "AI-generated" labels, nudges to pay to unlock intimate storylines, and preset characters heavy on scantily clad women with violent or extramarital plot options, despite April rules from the Cyberspace Administration and four other departments banning obscene, pornographic or violent content.

Parent sues Palo Alto schools over Turnitin AI cheating flag

Who did it: HYBRID Palo Alto Online

A parent has sued Palo Alto Unified School District in federal court, alleging a teacher wrongly accused his son of using AI on an essay after Turnitin flagged 76% of it, then made him rewrite it in class, dropping his grade to a C. The suit alleges discrimination against a multilingual Asian student, retaliation and lack of due process.

NC senator's suit over AI deepfake ad moves forward

Who did it: HUMAN Charlotte Observer

A federal judge in Charlotte let North Carolina state Sen. DeAndrea Salvador seek limited documents and up to five depositions from Whirlpool and Omnicom in her suit over an ad that used AI to twist her 2018 TED Talk into an endorsement of appliances sold by a Whirlpool subsidiary in Brazil. The companies argue North Carolina lacks jurisdiction.

Judge fines pro se Equifax plaintiff $500 for fabricated citations

Who did it: HUMAN AI Hallucination Cases database

Judge Susie Morgan sanctioned Taiujuan Burches, who is suing Equifax and others without a lawyer, $500 under Rule 11 on May 11, 2026. He had admitted fabricated citations in an earlier reply brief and was warned, then filed more inaccurate citations, and after the show-cause order cited a case the court could not locate.

Combs accuser's lawyer held in contempt over unpaid AI fine

Who did it: HYBRID Law360

Judge Leo M. Gordon held Tyrone A. Blackburn, lawyer for a woman suing Sean Combs, in contempt on May 11 for not paying a $6,000 fine imposed in December for citing a nonexistent, AI-generated case, but settled for an oral admonition rather than jail. The court has received no payments, the judge said; Blackburn said he sent one.

Singapore man who sent woman AI nude pleads guilty over spy cameras

Who did it: HUMAN CNA

Desmond Han Jiancong, 31, pleaded guilty on May 11 to installing equipment to enable voyeurism and insulting a woman's modesty. The court heard he sent the acquaintance an AI-generated topless image using her face and taped spy cameras under her desk for a live upskirt feed; no recordings were found. Sentencing is set for May 26.