Google Opens Cross-Vendor AI Watermark Detection: ChatGPT Images Can Also Be Checked

Beating AI News Flash: Google has opened SynthID Detector to the world. Users can upload images, videos, or audio to check whether the file contains an invisible watermark embedded when AI generated it. Previously, this tool was mainly available for testing by media professionals.

The new version of the detector also supports identifying SynthID watermarks from OpenAI, Nvidia, and Kakao, with Apple also set to join. In the past, Google's detector could only recognize its own watermarks, and even images generated by ChatGPT with SynthID could not be detected.

However, this tool can only identify supported SynthID watermarks and cannot detect all AI-generated content. Failure to detect a watermark does not mean the content was made by a real person.

The website requires login, and each user has about 10 detection credits per day.

Detection address: https://synthid.com