Google has expanded its SynthID technology to support watermarks from all its partners, making it easier for anyone to check if content is AI-generated.

Previously, the only way to verify an AI watermark was by asking Google’s Gemini model, which was a cumbersome process. Now, the new SynthID.com website allows users to check content without involving the model, providing a simpler ‘yes’ or ‘no’ answer.

With support for distinct watermarks from OpenAI, Nvidia, Kakao, and soon Apple, the expanded detector offers a more comprehensive check of AI-generated content. However, it will only provide a basic confirmation, rather than detailed information on the regions of the image or video affected.

The move signals a significant step towards transparency in the AI-generated content space, but also raises questions about the reliability of these watermarks and the extent of trust we should place in them.

Will this new tool truly revolutionize the way we verify AI-generated content, or will it become just another layer of tech complexity to navigate?