Advances in generative AI are enabling people to create content in completely new ways – from text to high-quality audio, images and videos. As these capabilities develop and become more widely available, questions of authenticity, context, and verification arise.
Today we're announcing SynthID Detector, a verification portal that enables you to quickly and efficiently identify AI-generated content created with Google AI. The portal brings discovery capabilities in one place across multiple modalities and provides essential transparency in the rapidly changing generative media landscape. It can also highlight which parts of your content are more likely to be watermarked with SynthID.
When we launched SynthID – a cutting-edge tool that embeds invisible watermarks and identifies AI-generated content – our goal was to provide a suite of innovative technical solutions to help minimize misinformation and misattribution.
SynthID not only preserves content quality, but acts as a robust watermark that remains detectable even as the content is shared or undergoes a series of transformations. While we initially focused solely on AI-generated images, we have since expanded SynthID to include AI-generated text, audio and video content, including content generated by our Gemini, Imagen, Lyria and Veo models at Google. Over 10 billion pieces of content have already been watermarked with SynthID.
How the SynthID detector works
When you upload an image, audio, video or piece of text created using Google AI tools, the portal will scan the media for the SynthID watermark. If a watermark is detected, the portal will highlight specific parts of the content that are most likely to be watermarked.
For audio, the portal indicates specific segments where a SynthID watermark has been detected, and for images, it indicates areas where a watermark is most likely to occur.
