SynthID
Google's watermarking system for AI-generated content including images, videos, and audio
Recently discussed in
What creators say
“The SynthID detector created by Google DeepMind to identify AI-generated or enhanced media by OpenAI, Google or others detected the images posted by Hidden and others appeared to be edited with AI image generation.”
“This morning, Google also launched SynthID. And SynthID is a system for watermarking proteins that are produced with AI... This is actually a pretty amazing technology that Google developed for creating watermarking and defense systems in AI bio.”
More on what David said about SynthID Hide
- Likes
- Embeds watermarks in both the DNA sequence and the 3D protein structure
“If you look at the Google SynthID, so Gemini has done this for like a few years. If you look at their blog that explains how it works, it will highlight specific words in the piece of text. It's like, okay, this is probably AI written. It's not saying the whole thing is AI.”
More on what Jordi said about SynthID Hide
- Good for
- Identifying AI-generated text at token level
- Likes
- Granular identification — marks specific words/tokens rather than whole document
- Compared with
- Compared to Anthropic's new watermarking feature
“Since launch, SynthID has now watermarked over 100 billion images and videos along with 60,000 years of audio assets.”
More on what Peter said about SynthID Hide
- Good for
- Watermarking AI-generated content for verification
Keep exploring
About this data
- ✓50,000+ mentions extracted from real podcast transcripts
- ✓Every quote links to the source episode
- ✓Sponsored mentions are always labeled
- ✓Critical takes are shown, never filtered out