Google Opens全球 Access to SynthID Detector AI Content Detector

• Nano Banana;
• Veo;
• Lyria;
• Gemini;
• Flow;
• ProducerAI;
• Vids.

SynthID Detector complements Google's verification features in Search, the Gemini app, and the Chrome browser. These tools process over a million queries per day. The technology is also used by Nvidia, OpenAI, and Kakao.

What can't the detector guarantee?

Google explicitly warns about its limitations. If the detector reports the absence of a watermark, this does not completely rule out that the content was AI-generated.

The mark may be missing for several reasons:

• the content was created by a model that does not use SynthID;
• the watermark was heavily altered;
• in rare cases, the system gives false answers.

Why does the market need this?

The main goal of the tool is to give people the ability to make more informed decisions when interacting with content created by artificial intelligence. Users get a tool for balanced assessment of materials: photos, videos, and audio are increasingly generated by neural networks, and distinguishing them from human-made content is becoming harder.

For the B2B sector — media, advertising agencies, and platforms — the detector becomes a practical tool for verifying content before publication or purchase. The scale of 180 billion watermarked files and support from Nvidia and OpenAI indicate that watermarks are becoming an industry standard for labeling generative content.