Google DeepMind Expands SynthID Detector Globally in English

Google DeepMind has expanded access to its SynthID Detector globally in English, allowing anyone to check if an image, video, or audio file was made with AI. Google introduced an early version of the SynthID Detector in 2025 for media professionals verifying AI-generated content. Now, the tool is available to everyone globally in English starting today.

The underlying watermarking technology, Google SynthID, was first launched in 2023. It helps people identify AI-generated media by applying imperceptible watermarks across images, video, and audio. Since that initial launch, Google has watermarked over 180 billion images and videos. The company has also applied these watermarks to 240,000 years of audio content.

With the expanded availability of the SynthID Detector, anyone can check whether an image, video or audio file was made with AI from Google or its partners. The supported partners currently include OpenAI, NVIDIA, Kakao, and Google plans to add Apple to this list soon.

The standalone detector joins Google's existing suite of built-in verification features. These features are already integrated into Search, the Gemini app, and Chrome, where they regularly handle over 1 million requests daily.