The Cyberspace Administration of China (CAC), the country’s leading internet regulator, has unveiled the country’s inaugural rules overseeing generative artificial intelligence. Hailed as an important regulatory step, these rules, which go into effect on August 15, were enacted as Beijing increased its vigilance over emerging technology.
These regulations oblige generative AI services to uphold the “core values of socialism”, highlighting the distinct socio-political background of AI development in China compared to the US
The global rise of generative AI.
Generative AI, which can generate text or images, is witnessing a global boom. Prominent US entities, including OpenAI’s ChatGPT, were at the forefront of this wave and pioneered a wave of similar services. CNBC reports that the rapid and widespread proliferation of these technologies has raised regulatory concerns worldwide.
China is no outlier in this trend, with CNBC noting that tech giants such as Alibaba and Baidu are actively refining their generative AI services. Still, the country’s unwavering control over the domestic internet and its monitoring of AI advancements have prompted careful strategy among these entities. As Reuters points out, Beijing’s increasingly strict technical regulations have slowed the public release of generative AI services, many of which are still in beta testing.
In the face of such challenges, the CAC’s rules are intended to facilitate the responsible development and use of generative AI. Under the new regulations, generative AI services, especially those accessible to the public, will require an operating license. CNBC reports that service providers who come across “illegal” content should stop creating it, fine-tune the algorithm and report the content to authorities. In addition, they must perform security assessments and secure user data.
China’s Generative AI Regulatory Provisions.
According to the CAC website, the core objective of the new regulations is to promote the healthy development and standardized application of generative AI, to safeguard national security and public societal interests, and to protect the rights and interests of citizens, legal entities and other organizations.
The CAC explicitly contains the following detailed provisions for the new regulatory mandate:
- Adherence to the core values of socialism, not generating content that incites harm to the state or promotes prohibited content such as violence, obscenity or false harmful information.
- Prevention of discrimination during AI algorithm design, data selection, model generation, optimization and service provision, based on characteristics such as ethnicity, religion, country, region, gender, age, profession, health, etc.
- Respect for intellectual property rights and business ethics, maintain trade secrets and avoid using algorithms, data or platforms to implement monopolies or unfair competition.
- Respect for the rights of others, ensuring that the AI does not endanger physical and mental health or infringe on individual rights such as image, reputation, honour, privacy and personal information rights.
- Implement effective measures to improve the transparency of generative AI services and improve the accuracy and reliability of generated content based on the service type.
Despite their apparent rigidity, these rules illustrate China’s delicate balancing act in fostering AI innovation while closely monitoring its progress. The CAC has stated that the regulation aims to “encourage innovative applications of generative AI and support the development of related infrastructure such as semiconductors”.
Criticism of the new regulatory mandate.
However, this seemingly progressive regulatory framework has drawn some criticism. Some experts, including Henry Gao, a law professor at Singapore Management University, told Reuters that China’s strict internet laws could potentially hinder the country’s ability to compete with its Western counterparts in technology. Gao argues that China’s “preemptive regulation” approach “could stifle innovation and slow down Chinese companies’ ability to catch up.”
Nevertheless, Reuters states that Chinese companies are moving forward and are mainly focusing on industrial applications of AI. This strategy is consistent with Chinese President Xi Jinping’s emphasis on more “hard” technology breakthroughs aimed at reducing China’s reliance on Western technology, an essential feature of the new regulatory landscape.
The CAC stated that at a time when the rapid development of generative AI technology has brought new opportunities as well as challenges such as the proliferation of false information, violation of personal information rights, data security issues and biases, the measures seek to balance the growth of generative AI and security and are designed to avoid the risks of generative AI services.
Details of the new mandate.
| mandate | Description |
|---|---|
| Socialist values | AI must uphold socialist values and not endanger state security or spread harmful information. |
| Non-discrimination | AI should prevent discrimination in all processes, including algorithm design and data selection. |
| Intellectual property rights | AI must respect intellectual property rights, keep trade secrets and avoid unfair competition. |
| Individual rights | AI should not infringe on rights such as health, privacy, reputation and personal information. |
| Transparency and accuracy | AI services must be transparent and ensure the accuracy and reliability of the content. |
| Security and reporting | Providers must conduct security assessments, protect user data, and report illegal content. |
| Stakeholder participation | Government, industry, society and internet users must participate in the development and governance of AI. |
The new measures propose to combine development with security, promote innovation and law-abiding governance, encourage the creation and growth of generative AI, and exercise prudent oversight of generative AI services.
The new specifications for generative AI services also require clarification of data processing and labeling requirements and require providers to take effective measures to prevent minors from becoming excessively dependent on generative AI services and to immediately address illegal content. In addition, there are systems for security assessment, algorithm submission, and complaint reporting, in addition to clear legal responsibilities.
China’s State Internet Information Office further emphasized that the development and management of generative AI services require the participation of government, enterprises, society and Internet users to jointly promote the healthy development of generative AI and ensure that the technology benefits the people good comes.
As the rapid evolution of generative AI continues, China’s pioneering regulations provide valuable insight for other countries struggling to oversee the nascent technology. Whether these rules will ultimately achieve the ideal balance of innovation and control remains to be seen. However, one thing is clear: the era of uncontrolled AI development is gradually disappearing in history.
