China makes first ChatGPT arrest – for manufacturing fake news.

Vice‘s Motherboard covers what is probably the first-ever AI-related arrest, by Chinese officials who accuse a man of using the software to mass-produce fake news:

The man, who is referred to by the surname Hong, used ChatGPT to write an article that claimed that nine people had died in a local train accident on April 25. This article was shared by more than 20 accounts on Baijahao, a blog-style platform run by Baidu, and had over 15,000 clicks by the time authorities’ spotted it.

Police in the Gansu province said that they detained Hong for “using artificial intelligence technology to concoct false and untrue information” and that he “exploited modern technological methods to fabricate fake information, and proceeded to spread it on social media.”

Investigators found that Hong would input elements of viral stories in China from the past few years into ChatGPT to produce different versions of the same fake story and then upload it to his accounts on Baijiahao. Though ChatGPT is not available in China, it can be accessed by using a VPN network. Hong, who operates personal media platforms, told investigators that he said that friends on WeChat had shown him how to make cash from clicks.


More at the link, and here, at the South China Morning Post.