← All stories
● Covered by 1 source · 1 reportMedium impact1 neutral

China Implements Interim AI Export Registration, Proposes New Frontier AI Controls

🔄 Updated 5d ago
New to BrevFeed? We gather this story from every outlet covering it into one summary — ranked by real-world impact, not just the latest headline — so you never miss what matters. What is BrevFeed? →

Key points

  • China issued Announcement No. 48 of 2026 on October 16, 2026.
  • An interim pre-export registration requirement is now in effect.
  • Draft proposes adding 23 AI technologies to export restrictions.
  • Controls target frontier model weights, agent architectures, and offensive cyber AI.

New Export Control Announcement

On October 16, 2026, China's Ministry of Commerce (MOFCOM) and Ministry of Science and Technology (MOST) released Announcement No. 48 of 2026. This announcement introduces two key measures: an immediate interim pre-export registration requirement for certain technologies and a draft proposal to expand the Catalogue of Technologies Prohibited or Restricted from Export.

Focus on Frontier AI Technologies

The draft additions to the catalogue include 40 technologies, with 23 specifically in artificial intelligence. Unlike previous AI-related entries in 2020, these new restrictions directly address frontier model weights, agent architectures, and AI-enabled offensive cyber capabilities. Explicit technical thresholds are defined for each restricted category.

Specific AI Restrictions

For model weights, restrictions apply if a model meets any of three criteria: at least 10^26 FLOPs of training compute, at least 2 trillion parameters, or a global top-10 ranking on public evaluations or comparable measures. Restricted components include weights, key checkpoints, source code, architecture, and training/inference optimization know-how. Agent technologies are also controlled, with restrictions on planning, multi-agent coordination, tool use, and game-theoretic decision technologies above set benchmarks. High-performance agents for critical infrastructure, defense, or core government use are outright prohibited.

Scope of Export and Enterprise Exposure

MOFCOM clarifies that access via API, app, web, SaaS, or lightweight on-device use does not constitute exporting a large model. However, deploying a restricted model locally outside China does count. Multinationals may also be affected through licensing deals, changes of ownership or control, and potentially intra-group code sharing. Any transfer of ownership, licensing for overseas use, or shift of control to a foreign entity can trigger the registration requirement.

Implementation Timeline

The interim pre-export registration requirement is already in effect, applying even to exports that have commenced but are not yet complete. The proposed catalogue additions are open for public comment until November 15, 2026.

✨ This summary was generated by AI from the outlets' reporting listed below. It is not independently verified and may contain errors — check the original sources. How BrevFeed works →

The daily brief

One email each morning: the day's tech stories, clustered across outlets and summarized. No account needed.

One email a day. Unsubscribe in one click, any time.

Today's brief

Spend a few minutes, get the whole day. Every topic's top stories in one hands-free rundown — listen, watch, or read the transcript.

~4 min · 3 stories · Nov 07

▶ Play today's brief Listen on Spotify

New every morning, and the back catalogue is archived by date.

Reporting from

China's Ministry of Commerce and Ministry of Science and Technology issued Announcement No. 48 of 2026, introducing an immediate pre-export registration requirement for certain technologies and proposing to add 23 AI technologies to its export restriction catalogue. These new draft restrictions target frontier model weights, agent architectures, and AI-enabled offensive cyber capabilities, marking China's first controls on advanced AI exports.