SynthID Detector — an innovative tool for recognizing AI-generated content
Sure! Here’s a rewritten version of the content with promotional references adjusted accordingly:
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 widely accessible, questions about authenticity, context, and verification arise.
Today, we’re introducing SynthID Detector, a verification portal designed to quickly and efficiently identify AI-generated content produced by Google AI. This portal offers detection capabilities across various media types in one location, providing essential transparency in the rapidly changing landscape of generative media. It also highlights which parts of the content are most likely to have been watermarked with SynthID.
When we launched SynthID — a cutting-edge tool that embeds imperceptible watermarks to identify AI-generated content — our goal was to create innovative technical solutions to help mitigate misinformation and misattribution. SynthID not only preserves the quality of the content but also acts as a robust watermark that remains detectable even when the content is shared or undergoes multiple transformations. Initially focused on AI-generated imagery, we have since expanded SynthID to encompass AI-generated text, audio, and video content, including work generated by our Gemini, Imagen, Lyria, and Veo models at Google. To date, over 10 billion pieces of content have been watermarked with SynthID.
How SynthID Detector works
When you upload an image, audio track, video, or text created using Google’s AI tools, the portal will scan the media for a SynthID watermark. If detected, it highlights specific portions of the content most likely to contain the watermark. For audio, it identifies segments where a SynthID watermark is found, while for images, it indicates areas where a watermark is most probable.
Let me know if you need any further adjustments!
