1 min read
Add as a preferred source on Google

Google Opens SynthID Detector Across Multiple AI Vendors

Users can scan images, video and audio for supported invisible watermarks from Google, OpenAI, Nvidia and Kakao. Apple is slated to join.

Hand holding a smartphone above three unmarked media cards / TokenPost.ai
Hand holding a smartphone above three unmarked media cards / TokenPost.ai

Google has globally opened its SynthID Detector, allowing users to scan images, videos and audio for invisible watermarks embedded during AI generation.

The detector recognizes supported SynthID signals from Google as well as OpenAI, Nvidia and Kakao. Apple is slated to join the supported vendors.

The expanded coverage means the tool can identify a ChatGPT-generated image when the file contains a supported SynthID watermark. Earlier access was limited to Google’s own watermarking system.

SynthID Detector does not identify every piece of AI-generated content. A file without a detected watermark may still have been created by artificial intelligence.

Users must sign in to access the website and receive about 10 checks per day.

Simon Yoon

Reporter

Simon Yoon reports on blockchain technology for TokenPost. Send corrections or tips to info@tokenpost.com.

Loading…