Man generates fake news using ChatGPT in China, gets detained
Authorities in the northwestern province of Gansu in China have detained a man for allegedly using conversational chatbot ChatGPT to create fake news and spreading it online, South China Morning Post reported. This is likely the first instance in the country where detention has been made related to using the artificial intelligence (AI) bot.
When the popularity of AI chatbots soared over the past few months, many expressed concerns over the misuse of the technology. Interesting Engineering recently reported an investigation that found more than 50 websites were generating content using AI, including that about celebrity deaths that never occurred.
China's internet regulator has also expressed concern about using synthesis technology such as deep fakes for defamation or online scams. It has put measures such as "clear labeling" of videos and photos made using the technology and placed in the public domain.
AI for generating fake news
As per the SCMP report, the officials of a county police bureau came across the fake news of a local train accident that claimed that the lives of nine people were lost. Upon further investigation, the article was found to be posted by more than 20 accounts on the blog platform called Baijiahao, which Baidu runs.
When the authorities noticed the news, it garnered over 15,000 clicks. They traced the story's origin to a company owned by the suspect, which operated personal media platforms in Southern China. During a search at Hong's home, the police found a computer and detained him.

As per his confession, Hong bypassed the duplication check on the platform by acquiring multiple accounts and publishing on them. He also admitted to using trending story elements in the past few years to create different versions of the same story using ChatGPT and then uploading them on the platform.
ChatGPT is not directly available to users in China, but it can still be accessed by those with a reliable VPN (virtual private network). Hong's suspected crime falls under "picking quarrels and provoking trouble," which could land him five years in prison. If considered a severe offense, the sentence could be extended to 10 years along with additional penalties.
Western social media platforms such as Twitter and Facebook have also introduced some measures to detect and prevent the spread of fake information created using AI technology. Still, it has not led to any arrests so far.