The site supports an extensive range of file types, including standard image formats like JPG and PNG, video containers such as MP4 and MOV, and common audio files like MP3 and WAV. Google originally debuted the technology at its I/O conference last year, initially limiting access to researchers and journalists. Now, the system integrates directly into the Gemini application and the Chrome browser, handling approximately 1 million verification requests daily.
Google Opens SynthID Tool to Public for AI Media Verification
Google has moved its SynthID verification technology out of restricted testing, launching a public website that allows users to scan images, videos, and audio files for AI-generated watermarks. The platform aims to bring transparency to synthetic media by identifying content created through the company's proprietary generative models.
SynthID embeds invisible digital watermarks into media produced by Google’s suite of generative tools, including Gemini, Veo, and Lyria. While industry peers like OpenAI, Nvidia, and Kakao have adopted the standard—with Apple reportedly planning to follow suit—the ecosystem remains fragmented. Microsoft and Meta continue to develop their own independent watermarking and detection protocols. Despite these efforts, detection tools remain imperfect, frequently struggling to identify content generated by their own creators' models.
Comments (0)
No comments yet. Be the first!