China is introducing new rules to restrict the use of deep synthesis technology.
This move is in line with the government's continuous push to combat internet rumors, propaganda, and misinformation, as reported by South China Morning Post (SCMP).
New Regulations
The Cyberspace Administration of China, the Ministry of Industry and Information Technology, and the Public Security Ministry issued the Administrative Provisions on Deep Synthesis for Internet Information Service on Dec. 11, which will go into effect next year on Jan. 10.
Any work that has been altered via deep synthesis must be clearly labeled so that it can be traced back to its original creators.
To adhere to the rules, you must get permission from the person whose picture or voice you are altering before doing so.
When reposting news with material made with deep synthesis tech, the source must be from the government-approved list of 1,358 sources.
The government has described deep synthesis as the use of software such as deep learning and augmented reality to produce text, pictures, audio, video, and virtual sceneries.
Those that provide deep synthesis services are expected to follow the law, act ethically, and keep the correct "political direction" and proper "public opinion orientation," as stated in the new legislation.
The Cyberspace Administration has said that they would provide strong legal protection to safeguard and support the orderly development of deep synthesis services by the implementation of these laws and regulations and defining systematic and tailored requirements for the use of such tech in online information services.
According to an article posted on the cyberspace administration's WeChat public account, as reported by SCMP, China's top internet regulator expressed worry that unregulated development and application of deep synthesis might lead to its use in illicit activities, including online fraud or defamation.
The decision to enact standalone legislation regarding deep synthesis is a response to the technological advancements and regulatory obstacles that have led to the current state of affairs.
Emergence of Deepfakes
Deepfakes, in which synthetic material is used to replace one person's face or voice with another, are one of the most infamous uses of the technology.
One notorious example is a doctored film purporting to show Facebook founder Mark Zuckerberg bragging about his command of stolen data from billions of Facebook users.
As technology improves, deepfakes are being utilized to fabricate celebrity porn movies, create fake news, and conduct financial fraud throughout the world.
Social media sites popular in the West, including Twitter and Facebook, have implemented new measures to identify and restrict the distribution of false news and propaganda enabled by deepfakes.
In the AI Sector
Meanwhile, developing deep synthesis technology presents problems and possibilities for the artificial intelligence (AI) business, according to Shi Jianzhong, vice principal of the China University of Political Science and Law.
Shi said that the advancement of the deep synthesis program and the expansion of its use cases would encourage the whole AI sector to promote the growth and ethical use of AI technologies.
Related Story : Deepfakes and AI: Soon to Change Our World? Samsung Creates its Version with 'MegaPortrait'
