SynthID Detector: A New Tool for Identifying AI-Generated Content

Here’s the revised content with neutral language and branding:

Advances in generative AI are enabling individuals to create content in entirely new ways — from text to high-quality audio, images, and videos. As these capabilities evolve and become more accessible, questions around authenticity, context, and verification arise.

Today, we’re introducing SynthID Detector, a verification portal designed to quickly and efficiently identify AI-generated content created with Google AI. This portal offers detection capabilities across various modalities in one location, providing essential transparency in the rapidly evolving landscape of generative media. It can also highlight which parts of the content are likely to have been marked with SynthID.

When we launched SynthID — an advanced tool that embeds imperceptible watermarks to facilitate the identification of AI-generated content — our goal was to provide a suite of innovative technical solutions to minimize misinformation and misattribution. SynthID not only maintains the quality of the content but also serves as a robust watermark that remains detectable even when the content is shared or undergoes various transformations. Initially focused on AI-generated imagery, we’ve since expanded SynthID to encompass AI-generated text, audio, and video content, including outputs from our Gemini, Imagen, Lyria, and Veo models within Google. Over 10 billion pieces of content have already been marked with SynthID.

How SynthID Detector Works

When you upload an image, audio track, video, or text created using Google’s AI tools, the portal scans the media for a SynthID watermark. If a watermark is found, the portal highlights specific portions of the content that are most likely to be marked. For audio, the portal identifies specific segments where a SynthID watermark is detected, and for images, it indicates areas where a watermark is most likely.

Feel free to let me know if you need further adjustments!