Tuesday, June 2, 2026

China Arrests Man Over Release of ChatGPT-Generated Fake News

Today in that’s-probably-not-how-you-use-artificial-intelligence, a man in China has been arrested for reportedly using ChatGPT to create a fake story about a train crash, Reuters reports

The perpetrator, identified only as a man from the Gansu province with the surname Hong, allegedly edited the news story generated by ChatGPT and then uploaded it to the Baidu-owned blogging platform Baijiahao.

The Pingliang city’s local police bureau said that the fake news claimed that a train crash killed nine construction workers in Gansu. It was posted on April 25 and received over 15,000 clicks on the platform. 

In 2021, there was a real train accident in Gansu that claimed the lives of nine people.

Reuters said that 25 Baijiahao accounts reported the fabricated incident, and Hong was identified and arrested after the police traced the IP addresses of all the accounts.

The incident marks the first time China has made an AI-related arrest as the capital begins to scrutinize the use of deepfake technology. In January, they enforced new rules that restrict service providers and users from accessing technology to create and distribute false information.

Hong has since been detained and investigations are ongoing.


Information for this story was found via Reuters, and the sources and companies mentioned. The author has no securities or affiliations related to the organizations discussed. Not a recommendation to buy or sell. Always do additional research and consult a professional before purchasing a security. The author holds no licenses.

Video Articles

SSR Mining Walks Away From a World Class Gold-Copper Project

Why More Canadians Are Starting to Think About Leaving | Jesse Day

Instead of Waiting, This Gold Developer Went Bigger | Kenneth McLeod – Sonoro Gold

Recommended

Nord Drills 61,389 g/t Silver Over 0.30 Metres at Castle East

Mercado Minerals Targets District Scale Silver Play With San Dimas Land Grab

Related News

Reddit Signs $60 Million Per Year Deal For AI Training Data

$60 million per year, so far — that’s how much licensing Reddit’s vast trove of...

Tuesday, February 20, 2024, 07:40:07 AM

AI Model Threatens Blackmail Over Affair in Company Safety Test

Anthropic‘s latest artificial intelligence model attempted to blackmail a fictional engineer over an extramarital affair...

Friday, July 4, 2025, 11:05:00 AM

Music Publishers File Preliminary Injunction Against Anthropic, Seeking to Halt ‘Improper Use of Copyrighted Works’

Three major music publishers – Universal Music, Concord Music Group, and ABKCO Music – have...

Tuesday, November 21, 2023, 03:44:00 PM

Almost 6 Months In, Is NVIDIA’s Eye Contact AI Still Too Creepy To Use?

Nvidia‘s (Nasdaq: NVDA) January release of Eye Contact, an AI-powered software video feature, garnered mixed...

Wednesday, July 12, 2023, 02:21:00 PM

OpenAI’s ChatGPT Reportedly In Talks For Tender Offer Putting Firm At $29 Billion Valuation

OpenAI, the research lab behind the ubiquitous ChatGPT chatbot, is in talks to sell current...

Friday, January 6, 2023, 11:55:00 AM