China's new regulation over Generative AI is less stringent than previously known
The country wishes to reap benefits from ChatGPT-like services while ensuring its political security.
A day after Premier Li Qiang sent a booster shot to Big Tech and the National Development and Reform Commission endorsed a number of their technology innovation projects - Artificial Intelligence in particular, China on Thursday released a much-more-supportive-than-expected regulation for Generative AI Services, or ChatGPT-like services in China, relaxing many restraints previously mentioned in an April Call for Comments draft.
The Interim Measures for the Management of Generative AI Services [CHN] (hereinafter referred to as the "Interim Measures") is jointly released by the Cyberspace Administration of China (CAC), National Development and Reform Commission, Ministry of Education, Ministry of Science and Technology, Ministry of Industry and Information Technology, Ministry of Public Security, State Administration of Radio and Television, and will come into effect on August 15, 2023.
It is significantly different from the Measures for the Management of Generative AI Services (Call for Comments) [CHN] (hereinafter referred to as the "Call for Comments") previously drafted by the CAC.
In a significant self-limiting clause, The Interim Measures stipulate that
"Industry organizations, enterprises, educational and scientific research institutions, public cultural institutions, relevant professional institutions, etc. that develop and apply generative AI technologies without providing Generative AI Services to the domestic general public are not subject to the provisions of these Measures."
That is to say, the measures regulate only "services provided to the general public" and do not regulate non-public-facing ChatGPT-like services. It echos the encouraging tone that the measures take towards scientific and technological advancements and opens up business opportunities for non-public-facing services. The clause was not available in the Call for Comments.
Article XI of the Call for Comments sets forth two conditions for companies BEFORE providing services by Generative AI Service products to the public:
First, to make a security assessment reporting to the CAC in accordance with the Provisions on Security Assessment of Internet Information Services with Public Opinion Attributes or Social Mobilization Capability.
Second, to complete algorithm filing procedures in accordance with the Internet Information Service Algorithm Recommendation Management Regulations.
The Interim Measures now say only the Generative AI Services “with the capacity to guide public opinion or mobilize society” should do so, and didn’t mention explicitly mandate they must do so BEFORE serving the public.
In a major step toward protecting the privacy of users of Generative AI Services, the Interim Measures have deleted a clause that requires the service providers to demand the real-name registration of users. So technically it now appears users can use the services anonymously.
Another small but probably not insignificant step is that the Interim Measures include a chapter of definitions of the terms used in the document, which is rare for Chinese government regulations and non-existent in the Call for Comments.
The Interim Measures add more burden on service providers in mandating the prevention of discrimination based on race, ethnicity, beliefs, nationality, region, gender, age, occupation, health conditions, etc. In the Call for Comments, only race, nationality, and gender were mentioned.
The gist of the changes between Call for Comments and Interim Measures is that Chinese decision-makers want to “balance development and security,” instead of over-regulating Generative AI Services and thus stifling potentially important progress from innovation.
That’s apparent in the broadening of the laws cited as the legal basis of the Interim Measures, which now includes the Law on Progress of Science and Technology in addition to the Cybersecurity Law, Data Security Law, and Personal Information Protection Law.
The Call for Comments draft was drafted and issued by the CAC alone, the Interim Measures were now jointly announced by CAC, the National Development and Reform Commission, Ministry of Education, Ministry of Science and Technology, Ministry of Industry and Information Technology, Ministry of Public Security, and State Administration of Radio and Television.
However, by Western standards, the Interim Measures would still appear to include a lot of politicized regulations, such as requiring the provision and usage of public-facing Generative AI services must
Adhere to the core socialist values, and shall not produce content that incites subversion of the state power or overthrow of the socialist system, endangers national security and interests, damages the national image, incites the division of the country, undermines national unity and social stability, promotes terrorism and extremism, spreads ethnic hatred and discrimination, violence, pornography, as well as false and harmful information, all of which are prohibited by laws and administrative regulations.
Will foreign providers of public-facing Generative AI services have a chance in the 1.4-billion-strong world’s second-largest economy?
If the provision of generative artificial intelligence services from outside the territory of the People's Republic of China to within the territory does not comply with laws, administrative regulations, and the provisions of these Interim Measures, the Cyberspace Administration of China shall notify the relevant institutions to take technical measures and other necessary measures to dispose of it.
Foreign investment in generative artificial intelligence services shall comply with the relevant laws and administrative regulations regarding foreign investment.
Here is a complete comparison between the Interim Measures and Call for Comments, in Chinese.
