China Moves to Extend AI Safety Rules Across the Full Lifecycle

China is accelerating legislation to put artificial intelligence development and use under safety oversight across the technology’s full lifecycle, a plan that would involve major model developers and address resources used to build AI systems.

National Development and Reform Commission official Wang Ruomeng said Tuesday that the effort would cover data and computing resources as well as AI development and applications. The standards are intended to reduce the risk that increasingly powerful AI systems become uncontrollable.

The regulatory effort would involve DeepSeek, 月之暗面 and Alibaba. The companies are developing products designed to compete with leading models from OpenAI and Anthropic.

The proposed approach ties safety oversight to multiple stages, from the resources used to develop AI to its eventual applications. The planned standards are described as among China’s most comprehensive AI safety guidance. The scope described links safeguards to both AI inputs and the later use of the resulting systems.