China GB/T 45654-2025 Basic Security Requirements for Generative AI Services
Clause 4: training data security (sources, content, annotation) – China GB/T 45654-2025 Basic Security Requirements for Generative AI Services

China GB/T 45654-2025 Basic Security Requirements for Generative AI Services 4.3.3: Accuracy of annotated content: batch sampling, re-annotation and review

What does 4.3.3 require?

For functional annotation the provider must manually sample every batch of annotated data, re-annotate a batch where content is found inaccurate, and discard a batch found to contain illegal or unhealthy information; for security annotation no item counts until at least one reviewer has checked and approved it.

Maintained by Gerard Blokdyk

Other controls in Clause 4: training data security (sources, content, annotation) – China GB/T 45654-2025 Basic Security Requirements for Generative AI Services

Query this from an agent

The graph holds this control, the 0 it maps to, and the evidence behind each claim, over MCP and REST.