China reports first arrest over fake news generated by ChatGPT

A man in China’s Gansu province has been detained for allegedly using ChatGPT to generate a fake story about a train crash, marking China’s first arrest in an AI-related probe as Beijing tightens deepfake technology. The story, which claimed the crash killed nine construction workers in a city in China’s northwestern Gansu, gained more than 15,000 clicks after being published on social media on April 25, Pingliang city’s local police bureau reported. China’s new rules for deepfake bar service providers and users from using such technology to produce, release and fabricate untrue information. The rules, which took effect from Jan. 10, are designed to curb the use of generative AI technology to alter online content. A total of 25 Baijiahao accounts, a blog-style platform owned by China’s tech giant Baidu, reported the “accident” with IP addresses from different locations, according to the police statement. It said the suspect, surnamed Hong, was identified and arrested for concocting false information after police traced all the accounts and found his company was involved. Hong allegedly edited ChatGPT-generated news and uploaded it on the Baidu-owned platform, local police reported, adding that investigations were continuing.