On Tuesday, a comprehensive new drafting regulation by China’s internet regulator revealed that future AI products developed in the country will now be subject to a security review before publication and must follow “core socialist values.”
Since San Francisco-based OpenAI debuted ChatGPT in November, triggering a gold rush in the market, a flurry of Chinese startups have rushed to develop AI services that mimic human speech.
With deep fake photos and people depicted mouthing things they never spoke, rapid breakthroughs in AI have raised concerns about the technology’s potential for misuse and spreading false information around the world.
According to the Cyberspace Administration of China’s draft law, a security assessment shall be applied through national internet regulatory departments before providing services to the public that utilize generative AI products.
The regulation that is being suggested under the term “Administrative Measures for Generative Artificial Intelligence Services” aims to ensure the healthy development and standardized application of generative AI technology.
The draft said that AI-generated content must reflect fundamental socialist ideals and cannot contain information that threatens the state’s legitimacy. Any components that can disturb social and economic order, such as content that promotes racial hatred or acts of terrorism, are also prohibited.
Given Beijing’s highly centralized political system, the Cyberspace Administration of China made the announcement that it was seeking public input on the new laws that were being suggested.