Advances in generative AI now allow us to create content in entirely new ways, from text to high-quality audio, images, and video. As these capabilities advance and become more widely available, issues of trust, context, and validation emerge.
Today, we’re announcing SynthID Detector, a verification portal for quickly and efficiently identifying AI-generated content created with Google AI. This portal provides detection capabilities across a variety of modalities in one place, providing essential transparency in the rapidly evolving produced media landscape. You can also highlight which parts of your content are likely to contain the SynthID watermark.
When we launched SynthID, a cutting-edge tool that enables the embedding of invisible watermarks and identification of AI-generated content, our aim was to provide a set of new technology solutions to help minimize misinformation and false attribution.
SynthID acts as a robust watermark that not only maintains the quality of your content, but also remains discoverable even when the content is shared and undergoes various transformations. Initially focused only on AI-generated images, we have since expanded SynthID to cover AI-generated text, audio, and video content, including content generated by Gemini, Imagen, Lyria, and Veo models across Google. More than 10 billion pieces of content have already been watermarked with SynthID.
How the SynthID detector works
When you upload an image, audio track, video, or piece of text created using Google’s AI tools, the portal scans your media for SynthID watermarks. If a watermark is detected, the portal highlights the specific parts of the content that are most likely to contain the watermark.
For audio, the portal pinpoints the specific segment where the SynthID watermark was detected, and for images, it shows the areas where the watermark is most likely to be present.
