AI in China: Key Regulations to Know

The document provides eight principles for AI governance:

1. Harmony and friendliness (AI should conform to human values, ethics, and morality, respect human rights, avoid misuse, and prohibit abuse and malicious application).

2. Fairness and justice (AI should promote equality of opportunity, eliminate bias and discrimination).

3. Inclusivity and sharing (AI should promote shared development, avoid data and platform monopolies, and encourage open and orderly competition).

4. Respect privacy (AI should protect personal privacy).

5. Secure/safe and controllable (AI should be transparent, explainable, reliable, and controllable).

6. Shared responsibility of AI developers, users, and other interested parties.

7. Open collaboration with full respect for each country’s principles and practices for AI governance.

8. Agile governance (AI should always moves in a direction that is beneficial to society).

The PIPL stipulates that personal information processing must be based on a legitimate purpose and with the consent of the data subject. Data processors must provide clear and explicit information regarding the purpose, method, and scope of data processing and obtain consent from individuals before collecting and using their data. 

Non-compliance with the PIPL can result in severe penalties, including fines of up to 50 million Yuan, or 5% of annual revenue, suspension of business operations, and criminal liabilities for severe violations.

The regulation encompasses several aspects, including content control and protection measures for workers influenced by algorithms, among others. It introduced the “algorithm registry,” which is also used in subsequent regulations on deep synthesis and generative AI. Service providers of algorithmic recommendation systems are required to promote mainstream values and transmit positive energy actively.

Moreover, the regulation includes provisions that prohibit providers from engaging in anti-competitive practices or excessive price discrimination through algorithms. It also forbids the development of algorithms that contradict ethical and moral standards, such as those that induce addictive behavior or excessive spending in users.

Individual users benefit from the regulation as well, gaining new rights, including the ability to disable algorithmic recommendation services, delete personalized recommendation tags, and receive explanations when an algorithm significantly affects their interests. The regulation aims to strike a balance between content control, consumer protection, and individual rights, promoting responsible and ethical use of algorithms in China.

The regulation requires that deep synthesis content conform to information controls, that it is labeled as synthetically generated, and that providers take steps to mitigate misuse. 

The regulation includes a number of unclear requirements, such as that deep synthesis content “adhere to the correct political direction,” not “disturb economic and social order,” and not be used to generate fake news. When such content “might cause confusion or mislead the public,” it must include a “conspicuous label in a reasonable position” to alert the public that it was synthetically generated. The regulation also includes a number of provisions targeting misuse.

The regulation mandates that generative AI should not be discriminatory on the basis of race or sex and that generated content should be “true and accurate”.

Generative AI services in China must also adhere to the “core values of socialism”. The rules will only apply to generative AI services that are available to the general public. Technology being developed in research institutions or intended for use by overseas users is exempted.

Generative AI services will need to obtain a license to operate.