China's economic planner says it will set AI safety standards covering the full life cycle of models
NDRC official Wang Ruomeng said China will speed up legislation on data, computing power and algorithms, with rules expected to cover firms such as DeepSeek, Moonshot and Alibaba.
Why it matters
PYMNTS, citing Bloomberg, says the measures would be among the most thorough guidelines anywhere aimed at preventing loss of control as AI grows more powerful. Wang said the rules would span data, computing power and algorithms, with safety standards and ethical guidelines across the full life cycle, regular reviews of how they work, and protections for workers displaced by AI. He also called AI safety a global challenge needing international cooperation. Watch for draft texts and which agencies enforce them.
Key facts
- Announced by: National Development and Reform Commission
- Official: Wang Ruomeng
- Scope: data, computing power and algorithms
- Covers the full AI life cycle
- Firms likely covered: DeepSeek, Moonshot, Alibaba
- Briefing date: 3 November 2026