China Deepfake Laws
The rules will enter into force on January 1, 2020 and will be enforced by the Cyberspace Administration of China. “With the introduction of new technologies such as deepfake in the online video and audio industry, there have been risks in using such content to disrupt social order and harm people`s interests, create political risks, and have negative implications for national security and social stability,” the CAC said in a note to online video hosting sites on Friday. according to the South China Morning Post. The regulations provide that people must be protected from imitation by deepfakes without their consent – images that are virtually indistinguishable from the original and can easily be used for manipulation or misinformation. On the page of the American platform, Facebook and Twitter are developing better tools to detect deepfakes and reduce the distribution of these videos and images on the respective platforms. Twitter said this month that after a series of high-profile incidents, including a deceptively edited video by House Speaker Nancy Pelosi that went viral, it drafted a deepfake policy that showed how vulnerable the company`s platform is to such misinformation. Last month, California became the first U.S. state to criminalize the use of deepfakes in the promotion and advertising of political campaigns. The law, called AB 730 and signed by Gov. Gavin Newsom, criminalizes the publication of audio, images or videos that give a false and harmful impression of a politician`s words or actions. California law doesn`t use the word deepfake, but it`s clear that AI-made fakes are the main culprit, as well as videos that have been deceptively edited to put someone in a negative light. [Deepfake tool/service providers must review user-created content. Internet companies have already needed thousands of content monitors that examine user-generated content, such as microblog posts and videos.
This can put serious pressure on a company`s profitability. However, failure to detect and remove sensitive content can result in serious penalties, including the complete shutdown of an app or platform] As elsewhere, deepfakes have been problematic in China. They have been used in financial scams, such as in 2020, when older women were deceived into thinking they had online relationships with celebrities. At the time, press reports mentioned that even though internet platforms were trying to detect deepfakes using technology, as required by legislation introduced earlier this year, they still needed human labor. Because deepfakes weren`t as heavily regulated as pornographic or violent content, they probably didn`t have the same priority. The new legislation will likely help by trying to solve the problem at its root: the tool or application in which deepfakes are created before they are even shared. While the European Union is still seeking advice from think tanks, China has taken steps to regulate AI and deepfakes. Interestingly, some of China`s 2020 regulations can also be found in the stoa report`s list of proposals. The Chinese government has unveiled a bill to crack down on deepfakes – the practice of using AI to turn existing digital content into realistic simulations of humans. Deepfakes are widely used in China – although they were not generally created for the purpose of political interference. In September, a new app called ZAO, which allowed users to share their faces with those of celebrities, garnered millions of views. Meanwhile, Chinese citizens also pay for deepfake videos that put the face of their choice – a celebrity or someone they know – on the body of a star.
Danning, Mr. Real names are now required for WeChat and other IMS. China Daily www.chinadaily.com.cn/china/2014-08/07/content_18268073.htm (August 7, 2014). Although China is the first country to introduce this type of policy, it is far from the only country trying to regulate manipulated media. Lawmakers in the United States, the United Kingdom, and the European Union have all begun to craft regulations and proposals to limit the power of algorithms and combat deepfakes, though existing protections for platforms and free speech would likely prevent laws as ambitious as China. But after a year-long crackdown on tech companies for everything from anti-competitive behavior to price discrimination, Beijing has shown a willingness to use regulations to wipe out billions of market capitalizations of the country`s largest tech companies in order to keep them in line with government goals. The tilt report mentioned above contains a detailed description (pages 221-237) in English by Bo Zhao on the regulatory framework for deepfakes in China at the time of publication (November 2021). On January 1, 2020, a law, the network audio-video information services Regulation (link in Chinese), came into force requiring platforms that produce and/or distribute audiovisual material to clearly label audio or video with deepfakes, deep learning, virtual reality or other new technologies. It also banned the use of deepfake and virtual reality (VR) technologies in the creation, publication or distribution of fake news and called on platforms to remove such media. Platforms also needed to set up easy-to-use complaint systems. The Cyberspace Administration of China (CAC), China`s main internet watchdog, has been tasked with enforcing the law. In March 2021, Chinese tech companies were asked to conduct a security review regarding deepfakes and compliance with the 2017 Cybersecurity Law.
Campbell, C. How China is cracking down on its once-untouchable tech titans. Time time.com/6048539/china-tech-giants-regulations/ (May 20, 2021). provisions on the management of online audiovisual information services (National Radio and Telecommunication Administration, 2019); www.chinalawtranslate.com/provisions-on-the-management-of-online-a-v-information-services/ Congress is also analyzing the potential damage of deepfakes and how best to combat their influence in the upcoming 2020 presidential election. The House Intelligence Committee held a hearing on the issue after convening a panel of experts from universities and think tanks to develop a deepfake strategy on electoral integrity and security. There are also many laws currently under consideration in Congress that would require special watermarks or disclosures regarding false or misleading media, as well as criminalizing the creation and distribution of such videos. Deepfakes are images, sounds or video recordings that are not real but seem very realistic. With the improvement of artificial intelligence (AI) technologies, deepfakes are becoming more and more realistic. Sometimes they are clearly made for entertainment purposes and are obviously wrong for everyone but the most gullible, like the video from Chinese AI company Iflytek in which Donald Trump speaks Mandarin.
But sometimes they are created to deliberately deceive the public. Chinese regulations take the Cybersecurity Law of 2017, the Data Security Law of 2021 and the Personal Data Protection Law of 2021, as well as several other laws, as the basis for new “deepfake” regulations. “The safeguarding of national security and the public interest of society and the protection of the legitimate rights and interests of citizens, legal persons and other organizations” are identified as objectives of the rules. Second, controlling the delivery of deeply synthesized content is inherently challenging. Real name verification laws can allow authorities to track sources and distributors of unlabeled or misleading in-depth synthetic content. However, providers of in-depth synthesis services are invited to immediately cease the transmission of unlabelled information and to initiate procedures to dispel rumours in order to combat disinformation and disinformation (Articles 15 and 17). This is a challenge because once a piece of content is created, it can be decoupled from the service on which it was created and distributed independently. Videos can be downloaded again, audio re-recorded, screenshots of images taken and removed from author control.
And as the saying goes, the Internet is eternal: once certain information has spread on the Internet, it is extremely difficult to remove it completely. In China, as elsewhere, content can be suppressed by censorship, but the collective memory can remain. For example, several weeks after the start of the COVID-19 lockdown in Shanghai, which began in March 2022, a video titled “âåæä声â (April Voice”), which recorded events during the lockdown through audio, began to spread on Weibo and reached millions of people through a relay of re-downloads and copies that were ahead of censors for a few hours.