China has just issued a number of regulations to strengthen control of "deep face-swapping" technology and prevent the use of artificial intelligence technology for impersonation.
The Cyberspace Administration of China (CAC) said the move is to limit the possible risks from the operation of platforms to change online content with deep learning technology (deep learning) or perform real-time actions. virtual reality.
The new rules will go into effect on January 10, 2023.
Deepfake is a technology that allows content providers to alter facial and voice data in the original with artificial intelligence-generated realistic images.
Deepfake technology can be used to illegally produce, copy, publish, disseminate information and even impersonate others to commit fraud and other illegal activities.
Deepfakes can even be weaponized and used to spread misinformation, mislead people, and rank among the top AI criminal threats.
China criminalized the use of AI to spread fake news in late 2019. At the time, the CAC raised concerns about the potential impact of deepfakes, leading many to speculate that the government would ban this technology altogether.
In doing so, however, China will miss out on important developments in the history of communications and artificial intelligence, as well as opportunities for global scientific cooperation.
Therefore, the state has decided to control rather than ban this potentially deceptive technology in order to encourage users to take advantage of it legally, especially in the entertainment sector.