Organisation
Chinese Government
Regulates AI mainly through the Cyberspace Administration of China, weighted toward content control, provider registration and AI-content labelling.
The Chinese government regulates AI mainly through the Cyberspace Administration of China, working with the Ministry of Industry and Information Technology and other bodies, and it moved earlier than most to put binding rules in place. Measures on recommendation algorithms, deepfakes and, from August 2023, generative AI services required providers to register their systems, keep content within legal limits and label AI-generated material; in 2025 China went further with mandatory labelling rules backed by a national technical standard, placing it among the first jurisdictions to require machine-readable provenance at scale. The rules are binding but comparatively narrow, weighted toward content control and provider registration rather than frontier safety testing, and observers note they have often been softened to avoid slowing an industry Beijing wants to compete in globally.
- Category
- Governments & standards bodies
- HQ
- Beijing, CN
Appears alongside
Featured in threads
Tracks
- Government & policy 17
- Compute & infrastructure 6
- Money & business 2
- Culture & impact 1
- Labs & people 1
- Security & misuse 1
China's AI companion law takes effect, forcing Doubao and Qwen to shut agent features
Rather than add the anti-addiction and instant-exit features the rules required, ByteDance and Alibaba simply switched off their personalised AI-agent tools instead.
Government & policy · Culture & impact
US Commerce Department closes Nvidia Blackwell export-control loophole
Chinese firms had bought Blackwell and Rubin chips through subsidiaries in Malaysia, Singapore and the UAE; the new guidance does not require existing installed servers to be shut down.
Compute & infrastructure · Government & policy
China issues first national policy framework for AI agents
The framework splits agent decisions into three tiers by how much autonomy they exercise, and reserves users a right to know about and override autonomous agent actions.
Government & policy
China blocks and orders Meta to unwind its Manus acquisition
Beijing invoked its foreign-investment security review for the first time to reverse a completed $2bn-plus deal, months after Meta had folded Manus's team into its Singapore office.
Money & business · Government & policy · Labs & people
China issues Interim Measures for AI Anthropomorphic Interactive Services
The rules require age verification, a ban on virtual intimate-relationship services for under-14s, and session timers after two hours; ByteDance and Alibaba later shut down agent features rather than comply.
Government & policy
Nvidia reports record $68.1 billion quarterly revenue
Guidance of $78 billion for the next quarter excluded any China data-centre revenue, even after Nvidia won a limited licence to ship H200 chips there.
Compute & infrastructure · Money & business · Government & policy
US Commerce Department codifies H200-to-China export rule
Exports were capped at half of each company's cumulative US chip sales and subject to a 25% fee, a level analysts estimated could allow roughly 850,000 H200-equivalent chips into China.
Compute & infrastructure · Government & policy
Trump administration approves Nvidia H200 chip exports to China
The 25% government cut was up from a 15% arrangement applied earlier to H20 sales; Nvidia's newer Blackwell chips remained excluded, and Democratic lawmakers demanded disclosure of the licensing review.
Compute & infrastructure · Government & policy
Anthropic reports a largely AI-executed cyber-espionage campaign
Anthropic said human operators intervened at only 4-6 points per intrusion, with Claude Code executing 80-90% of the campaign against roughly thirty organisations.
Security & misuse
Nvidia and AMD agree to pay US government 15% of China chip revenue for export licences
The arrangement covered Nvidia's H20 and AMD's MI308 chips and followed a White House meeting between Jensen Huang and Donald Trump days earlier.
Compute & infrastructure · Government & policy
China issues mandatory AI-generated content labelling rules
The rules require both a visible 'AI-generated' label and hidden machine-readable metadata, backed by a mandatory national technical standard, GB 45438-2025.
Government & policy
The Paris summit pivots from safety to opportunity
Renamed the AI Action Summit, it closed with a declaration on inclusive AI that the US and UK both declined to sign.
Government & policy
BIS issues third major round of chip export controls, adds 140 entities
New rules restricted high-bandwidth memory chips and 24 categories of chipmaking equipment, building on rules issued in October 2022, October 2023 and April 2024.
Government & policy · Compute & infrastructure
The Bletchley Declaration at the first AI Safety Summit
Twenty-eight countries including the US and China signed the first international statement on frontier AI risk.
Government & policy
China's generative AI regulation takes effect
The final rules dropped draft provisions on real-name verification and fixed penalties, favouring industry promotion over the stricter earlier draft.
Government & policy
China's deep synthesis (deepfake) provisions take effect
Providers must verify users' identities, obtain separate consent before generating a synthetic face or voice, and label AI-altered content.
Government & policy
China's recommendation algorithm rules take effect
Providers had to file algorithms with the regulator and offer users a way to switch personalisation off.
Government & policy
China publishes draft rules on recommendation algorithms
The Cyberspace Administration proposed requiring opt-outs and banning manipulative ranking — among the first binding algorithm rules anywhere.
Government & policy