The daily rap sheet of the machine age.
12 items
According to a Financial Times report, OpenAI and Google provided AI services to Singapore-based affiliates of Alibaba, Baidu and Tencent, whose parent companies are on the Pentagon's 1260H list of firms linked to China's military. The sales are legal under current US law but have prompted calls for tighter AI export rules.
Original report: OpenAI and Google sold AI to blacklisted Chinese firms' affiliates
The 11th U.S. Circuit Court of Appeals rebuked Florida lawyer and Lake County commissioner Anthony Sabatini for aviation employment case briefs citing nonexistent cases, AI-generated material it called "fake and hallucinated", and said he will be referred for discipline. "Whatever the merits of artificial intelligence, it is no substitute for actual intelligence," the panel wrote.
Original report: Appeals court rebukes Florida lawyer-commissioner over hallucinated citations
Federal District Civil Police served search warrants on two former employees of a private company suspected of using deepfake technology to put female co-workers' faces into fake pornographic videos and images, which police say were shared via corporate email and posted on an adult content platform. Phones and computers were seized in Gama and Paranoá.
Original report: Brasília police raid ex-employees over deepfake porn of co-workers
Kayseri police, working under the chief prosecutor, say driving-licence candidates who paid to obtain licences used ID cards bearing AI-generated photos merging their faces with other people's to get past exam security. Raids in six provinces detained 20; a court remanded 15 in custody on charges including forgery of official documents and released five under judicial control.
Original report: Turkish police bust AI face-merge driving licence exam ring
According to sources cited by EFE, Spanish-speaking scammers linked to a Chinese-origin criminal organisation have made about three million calls to Spain from Laos since October or November, using AI and caller-ID spoofing to pose as police in video calls and press victims to move money into "safe" accounts. How many calls succeeded was not specified.
Original report: Laos scam centres pose as Spanish police using AI
In preliminary findings from an investigation opened in May 2024, the European Commission told Meta to change Facebook and Instagram features it considers excessively addictive, including infinite feeds, highly personalised recommendations and video autoplay, or face fines of up to 6% of global annual turnover. Meta disputes the findings, citing its teen accounts.
Original report: EU tells Meta to change Instagram and Facebook features it deems addictive
Odisha's BJD, through its student wing, lodged an FIR at Bhubaneswar's Capital police station alleging a man posted an AI-generated deepfake video on Instagram on July 9 targeting opposition leader Naveen Patnaik and former bureaucrat V.K. Pandian. Seeking swift action, the party noted someone was already arrested over a deepfake of Chief Minister Mohan Charan Majhi.
Original report: BJD files police complaint over Patnaik deepfake video
In Dixon v. Cartagena, SDNY Magistrate Judge Jennifer Willis said attorney Blackburn, repeatedly sanctioned before for unverified AI citations, defended 17 non-verbatim "quotations" as "paraphrase or compression" and countered with accusations against Roc Nation that it says came partly from his own verification tool's hallucinations. The court supplied dictionary definitions of quotation marks.
Original report: Judge explains quotation marks to lawyer defending inexact quotes
A research report, based on interviews in northeast Nigeria with 27 former Boko Haram members, says both factions used ChatGPT, Claude, Gemini, Grok, Meta AI and DeepSeek for attack planning, weapons troubleshooting and explosive device design, circumventing some safeguards, aided by specialised units and Islamic State trainers. Documented use remained conventional.
Original report: Boko Haram uses frontier chatbots to plan attacks, study finds
Days after launch, Meta removed a Muse Image feature that let users generate AI images by @-mentioning public Instagram accounts, without alerting the people whose photos were used. The company said the feature "missed the mark"; Puck News reported the reversal came amid scrutiny from users and talent agencies, including CAA.
Original report: Meta pulls Instagram AI tool that borrowed strangers' public photos
California's First Appellate District ordered attorney Carlton Floyd to pay sanctions to the court for an opening brief that misstated case law and included fictional quotations, in a fee dispute arising from a dairy succession contract. The court otherwise partly sided with his client, reversing the exclusion of paralegal fees, and rejected claims the appeal was frivolous.
Original report: California appeals court sanctions lawyer over fictional brief quotations
An influencer accused Nollywood actress Anita Joseph of publicly posting the influencer's photo with Joseph's face swapped in. Joseph apologised, saying her "AI boy" called it "an inspiration from AI" and suggesting a member of her creative team made it without her knowledge; the influencer replied: "SHE STOLE IT."
Original report: Nollywood actress blames her 'AI boy' for face-swapped photo
