In March of this year, the State Internet Information Office, the Ministry of Industry and Information Technology, the Ministry of Public Security, the State Administration of Radio and Television formulated the "Artificial Intelligence Generation of Synthesized Content".markingsThe Measures shall come into force on September 1, 2025 (yesterday).
And as required by the AI Generated Synthesized Content Labeling Scheme.Multiple platforms have announced the launch of AI-generated synthetic logos to help users better identify if content is made by AI-generated:
DeepSeek: On September 1, it announced that it had added logos to AI-generated synthetic content on its platform, and explicitly reminded users that the content in question was generated by AI.
Jitterbug: On September 1, Jitterbug announced that it had launched two core functions: AI content identification function, which assists creators in adding cue identifiers for AI content to make it easier for users to differentiate between them; and AI content metadata identification read/write function, which identifies and writes metadata information about AI content to provide technical support for content traceability. Meanwhile, Jitterbit will verify and detect whether the content in the station that is not actively labeled is generated by AIGC technology, and add explicit labels to works that are suspected to be AI content or whose metadata already contains AI content labels.
Weibo: By upgrading product functions and technical identification means, implicit and explicit logos are added to AI-generated content identified on the platform in accordance with the rules; if the text, images, audio and video content published by users is generated and synthesized by AI, the user should select "content generated by AI" in the "Select Content Statement" function column. In addition, "No AI logo added" has been added to the complaint side,
Fast Hand: The platform has already launched the AI content logo, and will add the "AI generated" explicit logo and the implicit logo embedded in the metadata of the file to the AI-generated content. If the content uploaded by users is clearly generated by AI or suspected to be generated by AI, the platform will provide a prominent reminder around the content.
Tencent: For AI-generated and synthesized text, images, audio, and video content obtained through the platform, the platform will add a clear logo of "AI-generated" or "AI-synthesized" in a prominent position; the platform will also add implicit logos in AI-generated and synthesized content through metadata embedding and protection technologies.
KunlunWanwei: The platform has added explicit watermarks and implicit metadata identifiers with the words "AI Generation" to all AI-generated interaction scenes and AI generation results; when creators publish or secondary disseminate AI-generated content through this platform, they must add the topic label "#AI Generation #" in a prominent position such as the title or the first screen and clearly state in the introduction the model and version number used. model and version number used.
It is reported that the "marking approach" makes it clear that AI-generated synthetic content marking mainly includes explicit marking and implicit marking in two forms:
Explicit logos are logos that are added to the interface of the generated synthetic content or interaction scenarios, which are presented in the form of text, sound, graphics, etc. and can be clearly perceived by the user;
Implicit marking refers to technical measures taken to add marking to the data of the generated synthetic content file that is not easily and obviously perceived by the user.