Documento :

UIT-D SG02 RAPPORTEUR GROUP MEETING  Contribución  81

Título :

Adoption and utilization of new and emerging AI technologies for content authenticity and identity security

Recibido en :

2026-09-24

Origen :

China (People's Republic of)

AI/Cuestión :

Q5/2

Reunión :

2026-10-07

Disponibilidad :

Restringido a usuarios de TIES [UIT-D]

Abstracto :

The rapid proliferation of generative AI has driven content tampering and identity forgery into large-scale abuse, threatening the trustworthiness of digital services such as finance and government. The international community has already built a common foundation through ITU and other standards bodies - under the World Standards Cooperation (WSC) framework, the AI and Multimedia Authenticity Standards (AMAS) collaboration, jointly launched by ITU together with IEC and ISO, is advancing standards for AI watermarking and multimedia authenticity. Building on this foundation, the China Academy of Information and Communications Technology (CAICT), together with domestic industry associations and academic societies, launched the "HuZheng" (Identity and Credential Protection) Initiative at the 2024 World Internet Conference, promoting the building of credible anti-forgery capabilities for identity documents under the principle of "using AI to guard against AI". CAICT led the development of the group standard "Technical Requirements for Text Image Tampering Detection Systems", published in September 2024, and is currently drafting the "Specification for Preventing Intelligent Identity Forgery in Authentication" (draft for comments). This contribution presents the above work and proposes areas for international cooperation, including the development of digital skills.