Google opens SynthID Detector to all users for AI watermark checks
The tool at SynthID.com flags watermarks in images, video and audio, including content from OpenAI, NVIDIA and Kakao, with around 10 checks per day.
Google has made its SynthID Detector available to everyone, allowing users to upload an image, video or audio file and check whether it carries a SynthID watermark. The tool is accessible at SynthID.com, according to Engadget.
The detector also identifies watermarks in content produced with models from SynthID partners OpenAI, NVIDIA and Kakao. Support for detecting Apple-created content is still in development.
Use of the tool requires signing in with a Google, OpenAI or Apple account. Users are limited to around 10 checks per day, a cap Google says is intended to prevent attempts to remove SynthID.
The detector only reports whether a label is found; it does not show which parts of a file were AI-generated. It also cannot detect media made with AI models that do not use SynthID, such as Meta's tools.
Earlier rollout
Google introduced SynthID in 2023 as an imperceptible label added to AI-generated content. In 2025, the company launched a limited version of the detector for certain researchers and journalists.
Quick answers
Where can I use Google's SynthID Detector?
The tool is available at SynthID.com and requires a Google, OpenAI or Apple account to sign in.
How many files can I check per day?
Users are limited to around 10 checks per day.
Can it detect content from OpenAI, NVIDIA or Kakao?
Yes. The detector can identify watermarks in content made with models from SynthID partners OpenAI, NVIDIA and Kakao. Detection of Apple-created content is still in development.