The South China Morning Post (opens in new tab) reports that Chinese police have arrested a man who allegedly used ChatGPT to create a fake news story and spread it online. This may well be the first instance of a person being detained for their use of AI, though it certainly won’t be the last.
The man was detained in the northwestern province of Gansu and was identified by the police using only his surname, Hong. In a statement the force said he’d been held for “using artificial intelligence technology to concoct false and untrue information.”
The article in question appeared on April 25 and falsely claims that there had been a local train crash in which nine people had been killed. Cybersecurity officers found that the article had been simultaneously posted across more than 20 accounts on Baijiahao, a microblogging platform run by Chinese search engine giant Baidu, and been read by at least 15,000 people.
It’s the first arrest since China introduced new laws to regulate the use of AI and ‘deepfake’ technology in January. The Administrative Provisions on Deep Synthesis for Internet Information Service target any technologies that generate text, images, audio or video, and explicitly mention deep learning models. It doesn’t outlaw the creation of things using these technologies, but does force them to be “clearly labelled” as such.
The police traced the article to a company owned by the suspect, and 10 days after it appeared the police arrested Hong, took his computer, and searched his house. The police statement says Hong confessed to using elements from past trending stories in China to input into ChatGPT, producing various versions of the story quickly, before they were uploaded to the Baijiahao accounts. Hong claimed he’d done it after friends on WeChat told him how to make money from clicks.
Hong’s crime is, per the police, “picking quarrels and provoking trouble”, which carries a maximum sentence of five years. However authorities can choose to make an example and, given this is a first, Hong may well be in very hot water.
This is all part of a wider and ongoing tech crackdown from Beijing, as the authoritarian Chinese Communist Party seeks to exert its control over emerging tech and how China’s population uses it. The Chinese police issued a statement in February warning the public to be wary of rumours created by ChatGPT.
GPT’s reaction to this article:
As an AI language model, I do not have opinions, but I can provide information. The article reports on the first arrest made in China due to the use of AI to spread fake news. The man allegedly used ChatGPT to create a false news story and spread it online. This incident highlights the Chinese government’s efforts to regulate the use of AI and deepfake technology and control the spread of fake news. The new laws in China do not outlaw the creation of things using these technologies but require them to be “clearly labelled” as such. The suspect may face a maximum sentence of five years for his crime of “picking quarrels and provoking trouble.”