China: TC260 publishes four guidelines on AI application security

《人工智能安全治理框架3.0》TC260 published four guidelines on AI application security on September 15, 2026

https://www.tc260.org.cn/tc260/xwdt1/202609/e879077a3caa4722b2206d1bcaed5a6c/files/%E3%80%8A%E4%BA%BA%E5%B7%A5%E6%99%BA%E8%83%BD%E5%AE%89%E5%85%A8%E6%B2%BB%E7%90%86%E6%A1%86%E6%9E%B63.0%E3%80%8B.pdf.pdf

• General Principles: Guidelines cover planning, risk classification, design, development, verification, deployment, operation, and decommissioning.

• Sector-Specific Guidance: Guidelines include specific recommendations for education, healthcare, and broadcasting and online audiovisual services.

• Personal Information Security: Emphasis on obtaining consent, conducting impact assessments, and ensuring data security.

• Human Oversight: Recommendations for human intervention measures, including emergency shutdowns and version rollbacks.

Things to Consider

• Implement AI security measures: Organizations should establish clear accountability, conduct annual security testing, and reassess systems when risks or application scenarios change.