Home Page >  News List >> Tech >> Tech

Using AI to fabricate fake news about "train crashing into people"! China's first ChatGPT crime, Dongguan man may face 5 years in prison

Tech 2023-05-18 03:17:57 Source: Network
AD

On April 25th, according to a report on the Baijia train, a train in Gansu Province collided with road workers, resulting in 9 deaths.The police immediately launched an investigation into this and found that a total of 21 accounts had simultaneously published different versions of articles on the platform, involving different accident locations, including Kongtong District, Lanzhou, Longnan, Dingxi, Qingyang, and other places in Pingliang City

On April 25th, according to a report on the Baijia train, a train in Gansu Province collided with road workers, resulting in 9 deaths.

The police immediately launched an investigation into this and found that a total of 21 accounts had simultaneously published different versions of articles on the platform, involving different accident locations, including Kongtong District, Lanzhou, Longnan, Dingxi, Qingyang, and other places in Pingliang City.

A train carrying passengers in the Netherlands collided with a crane near The Hague on Tuesday and derailed, causing at least one worker to die and about 30 people to be injured. This is the most serious railway accident in the country in many years.

The relevant reports quickly gained attention and received over 15000 clicks in a short period of time.

Think about it, have similar accidents occurred multiple times at the same time? This is too coincidental.

Facts have proven that these reports are indeed fake news.

After investigation by the Cybersecurity Brigade of the Chinese Public Security Bureau, it was found that the accounts that released the reports were all affiliated with a self media company in Shenzhen, Guangdong.

On May 5th, the police went to the suspect's residence in Guangdong for evidence collection. After interrogation, it was found that the man surnamed Hong first searched for widely discussed social news in China in recent years, and then used ChatGPT to modify specific elements such as time, location, date, or gender, and rewrite them into a fake news article. By uploading these articles, the account earns traffic and then monetizes it.

Not only that, the man surnamed Hong also used ChatGPT to create multiple fake news versions with small gaps to avoid censorship on Baidu's platforms.

It is understood that the man surnamed Hong mainly made illegal profits by modifying and editing the news elements collected through ChatGPT and uploading them to the Baijia account he purchased. The case is currently under further investigation.

The police stated that the behavior of a man surnamed Hong who fabricated fake news and spread it on the internet for people to browse extensively has been suspected of provoking and causing trouble. Therefore, criminal coercive measures have been taken against him, with a maximum sentence of five years in prison.

The first case of being arrested for abusing artificial intelligence

This is the first incident that China has detected since the promulgation of the Administrative Provisions on the Deep Synthesis of Internet Information Services in January this year in response to the development of artificial intelligence.

The regulation aims to promote the development of the artificial intelligence industry and prevent technology abuse. Synthetic videos and photos produced through deep fakes must be clearly labeled to prevent public confusion.

The regulation also covers the responsibilities of deepmake providers and users, including prohibiting the use of the technology for illegal activities, establishing censorship systems, and identifying user information.

Recently, with the popularity of ChatGPT, Chinese online search giant Baidu and other technology companies have also launched competing products.

On April 11th, the National Internet Information Office (CAC) of China released a draft regulation for managing generative AI services. According to CAC, the regulation is expected to take effect at some time this year.

ACA stated that the country supports independent innovation and promotion of artificial intelligence applications, but related products or services should comply with legal and regulatory requirements, respect social morality and public order, including service provider access, algorithm design, training data selection, content summarization, user privacy, and trade secrets.

Liu Xiaochun, associate professor of the Law School of the University of the Chinese Academy of Social Sciences, said in an interview that service providers are obliged to take risk prevention measures and respond to regulatory requirements.

Standardizing the use of AI is a global challenge

Using AI such as ChatGPT to generate fake news is not unique to China. American AI experts, social media companies, and government officials are equally concerned about the possibility of generative AI spreading erroneous information.

In April of this year, the Republican Party of the United States released an artificial intelligence generated video advertisement, assuming that Biden's re-election would trigger a massive wave of immigration along the US Mexico border and the collapse of San Francisco civilization.

Although this advertisement is labeled as "generated by artificial intelligence," the overly realistic image still makes it easy for people to see it as reality. The Midjournal sub section on Reddit also has users providing examples of how text to image generators can create historical events that have not occurred before, such as the "2001 Cascadia 9.1 earthquake and tsunami".

In addition, AmnestyNorth, a branch of human rights non-profit organization AmnestyWorld, has posted artificial intelligence generated images on social media accounts, calling for more attention to the two-year national strike in Colombia. But these images mistakenly depict the Colombian flag and use dummies instead of real protesters.

With the increasing attention and use of ChatGPT, domestic legislators have been working hard to draft relevant regulations on how companies can develop artificial intelligence. The proposed regulations include not generating false information, clearly marking content produced using AI, and ensuring that data used for training AI does not discriminate against race, race, and gender.

But it can also be imagined that this is a challenging and lengthy task that requires time to gradually polish.


Disclaimer: The content of this article is sourced from the internet. The copyright of the text, images, and other materials belongs to the original author. The platform reprints the materials for the purpose of conveying more information. The content of the article is for reference and learning only, and should not be used for commercial purposes. If it infringes on your legitimate rights and interests, please contact us promptly and we will handle it as soon as possible! We respect copyright and are committed to protecting it. Thank you for sharing.(Email:[email protected])

Mobile advertising space rental

Tag: Using AI to fabricate fake news about train crashing

Unite directoryCopyright @ 2011-2024 All Rights Reserved. Copyright Webmaster Search Directory System