On 1 August 2025, the National Information Security Standardisation Technical Committee (TC260) issued the new Framework for Artificial Intelligence Computing Platforms (GB/T 45958-2025). The standard will enter into force on 1 February 2026, and it is applicable to all stages of the design cycle of artificial intelligence computing platforms. The standard establishes a security framework for artificial intelligence computing platforms and specifies security functions, security management, and role security responsibilities. The standard aims to prevent risks on AI computing platforms, promote trust in AI technology, and provide a security foundation for future development.
Original source