U.S. District Judge Vince Chhabria let Kadrey v. Meta move forward, ruling that authors including Richard Kadrey, Sarah Silverman and Ta-Nehisi Coates adequately alleged Meta intentionally removed copyright management information to conceal infringement when training its Llama models on their books. He dismissed their claims under California's computer data access and fraud law. Meta says its training is fair use.
A student in Yale's executive MBA program, suing as John Doe, says he was suspended for a year for misleading the Honor Committee after being wrongly accused of using AI on an exam. The Feb. 3 suit says an instructor ran some of his answers through 'ChatGPTzero' after a teaching assistant flagged near-perfect grammar, and alleges discrimination against him as a non-native English speaker.
A CSIS report says Huawei, which is on the US entity list, used shell companies to have TSMC produce more than two million Ascend 910B logic dies in violation of US export controls, enough for about one million 910C chips. A Taiwanese official conceded the chips 'ended up with Huawei', it says, adding that Huawei also stockpiled Samsung memory via shell companies.
Jinan Art Museum in Shandong displayed an AI virtual figure, 'Jinan Xiao Ni'er', with a girl's face but an adult woman's body that twists and poses during interactions. Netizens accused it of sexualising children and demanded its removal. The museum said it had received no complaints and that the work had passed strict review before display.
