China: TC260 adopted Guidelines for Detecting AI-Generated Synthetic Content

On 28 August 2025, the National Technical Committee on Cybersecurity Standardisation (TC260) adopted the Cybersecurity Standard Practice Guide Detection of Artificial Intelligence Generated Synthetic Content Part 1: Framework, providing a structured technical framework to implement Section 6 of the Administrative Provisions on the Labelling of Artificial Intelligence Generated Synthetic Content. The Guide specifies objectives, detection processes, algorithms, and service encapsulation requirements to standardise the identification of synthetic content generated through artificial intelligence (AI) technologies across multiple modalities, including text, images, audio, and video. It defines general and specific detection goals, including accuracy, adaptability, resistance to distortion interference, localised detection, real-time application, and algorithm attribution. The Guide outlines end-to-end detection procedures comprising scheme design, preprocessing, implementation, and output, incorporating algorithmic approaches such as statistical models, multimodal detection, adversarial training, and fingerprint analysis, and further sets out normative requirements for encapsulating detection services in Software Development Kits (SDKs), Software as a Service (SaaS) deployment, and standalone products. Developed in alignment with GB/T 25069-2022 Information Security Technology — Terminology and GB 45438-2025 Cybersecurity Technology-Identification Methods for AI-Generated Synthetic Content, the Guide establishes technical specifications intended for use by online content dissemination service providers and third-party evaluation institutions in ensuring compliant detection, reporting, and integration of AI-generated synthetic content within regulated environments.