China Enhances AI Oversight. Prioritizing Safety and Data Protection
China’s top legislature is enhancing regulations concerning artificial intelligence safety and ethics with a proposed amendment to its cybersecurity law, as per reports. The decision underscores Beijing’s increasing emphasis on AI governance in light of escalating worries regarding misinformation, privacy violations, and online fraud. On Thursday, Wang Xiang stated that the amendment draft was initially reviewed last month. Lawmakers concurred that it would act as a “guiding principle for work related to internet safety” and incorporate new sections on AI regulation and personal data protection.
Wang stated that since the cybersecurity law was first enacted in 2016, the rapid rise of new technologies has introduced new risks and illegal online activities. The amendment seeks to address these challenges by establishing a framework for the security and development of AI. The proposal encompasses backing for research in essential AI theories, pivotal technologies, and foundational infrastructure. The emphasis is also placed on elevating ethical standards, bolstering risk monitoring, and augmenting oversight of safety related to AI. In China’s legislative process, a draft law generally goes through three reviews before it is passed; however, it may be approved earlier under specific conditions. The NPC Standing Committee is set to discuss the amendment during its session, which will commence on Friday and conclude on Tuesday.
The recent legislative initiative comes in response to a sequence of actions taken by the Cyberspace Administration of China aimed at regulating AI-generated content. In March, the CAC introduced regulations mandating both visible and embedded labels for AI-generated text, images, audio, and video. Labels that are visible should be unmistakably clear to users, whereas hidden identifiers, like digital watermarks, need to be included in the metadata. In 2023, the CAC issued regulations requiring that providers of generative AI services safeguard personal data, guarantee information accuracy, and uphold intellectual property rights. Despite these regulations, the misuse of AI persists, prompting ongoing concerns.
As China emphasizes regulation, analysts foresee that its AI industry may soon emerge as a significant source of revenue. Reports says that Chinese companies could begin to see profits from AI-powered “agents” as soon as next year, with 2026 anticipated to be the year of agent monetisation. UBS estimates that the US currently leads the AI agent market, generating between $15 billion and $20 billion in annual revenue, with American firms having a competitive edge due to their familiarity with enterprise-level software and AI tools. Chinese companies have concentrated on AI applications that engage consumers, particularly in e-commerce and entertainment. However, major technology companies such as Tencent, Alibaba, and ByteDance are currently competing to develop their own “agentic frameworks,” toolkits designed to empower AI agents. While US frameworks continue to hold a dominant position, Chinese alternatives are increasingly gaining traction, with IBM rankings indicating that ByteDance’s Coze Studio and Alibaba’s Qwen-Agent have each exceeded 10,000 stars on GitHub, showcasing their growing popularity.
