Beijing Issues 3rd AI Safety Framework as Chinese Leaders Dismiss U.S. Risk Warnings
Updated
Updated · CNBC · Sep 16
Beijing Issues 3rd AI Safety Framework as Chinese Leaders Dismiss U.S. Risk Warnings
3 articles · Updated · CNBC · Sep 16
Summary
China released the third edition of its AI Safety Governance Framework during cybersecurity week, setting guidelines for labeling AI-generated content and building rapid risk-detection systems.
Chinese executives and officials cast recent U.S. calls to slow AI development as overblown or self-serving, with the foreign ministry calling them fear-mongering and startup founders describing them as marketing.
That stance reflects a different industry focus: Chinese firms have prioritized commercialization and enterprise use under earlier state controls, while U.S. labs have pushed harder toward AGI despite rising scrutiny in Washington.
Beijing already vets generative AI services, restricts politically sensitive outputs and is folding AI adoption into its five-year economic plans, while promoting open-source models and AI cooperation through BRICS and other initiatives.
Chinese companies still see hallucinations and weak near-term profits as the bigger practical problem, often relying on older or smaller models to deliver productivity gains rather than chasing the most advanced systems.