Rules / China

GB 45438-2025 Cybersecurity technology – Labelling method for content generated by artificial intelligence

In forceStandardBinding网络安全技术 人工智能生成合成内容标识方法（GB 45438-2025）

Mandatory national standard released alongside the CAC labelling measures, specifying how explicit labels and implicit (metadata) labels must be implemented for AI-generated text, images, audio, video and virtual scenes.

Why it matters

As a mandatory (GB, not GB/T) standard it sets the technical test for whether labelling is compliant.

What it requires

Labelling AI content
SourceCAC (Q&A on the Labelling Measures): https://www.cac.gov.cn/2025-03/14/c_1743654685896173.htmSecondary: https://openstd.samr.gov.cn/bzgk/gb/std_list?p.p1=0&p.p2=45438 Checked against the source on 4 Oct 2026. Issued 28 Feb 2025 and publicly released with the CAC Labelling Measures on 14 Mar 2025; implementation 1 Sep 2025 (SAMR national standards database lists GB 45438-2025 as a current mandatory standard, issued 2025-02-28). CAC Q&A confirms the standard and the Measures took effect together.

Cite this record

GB 45438-2025 Cybersecurity technology – Labelling method for content generated by artificial intelligence. State Administration for Market Regulation (SAMR) and Standardization Administration of China (SAC); proposed by the Office of the Central Cyberspace Affairs Commission. Status: In force. wheresthe.ai, https://wheresthe.ai/rule/cn-gb-45438-2025-aigc-labelling-standard/ (verified 4 Oct 2026).

Line of Thought

Developments connected to this rule

What has happened that this rule shapes, or that shapes it. Pick one to keep following the thread.

More from China