synthid-text

This is interesting, is there a version with shorter responses? I spent a few minutes reading about SynthID-Text [0] and couldn't find mention of this (I've not read the paper yet), but my intuition [1] is that encoding more bits into a shorter piece of text would necessarily require a more noticeable transformation. Curious to see what that looks like, but don't have time to spin up my own version of this right now. [0]: https://github.com/google-deepmind/synthid-text [1]: my intuition .. which could be incorrect, that's why I want to see :)

There's layers, some of token selection is fingerprinting https://github.com/google-deepmind/synthid-text

Not foolproof, but a couple of easy ways to verify if images were AI generated: - OpenAI uses the C2PA standard [0] to add provenance metadata to images, which you can check [1] - Gemini uses SynthId [2] and adds a watermark to the image. The watermark can be removed, but SynthId cannot as it is part of the image. SynthId is used to watermark text as well, and code is open-source [3] [0] https://help.openai.com/en/articles/8912793-c2pa-in-chatgpt-... [1] https://verify.contentauthenticity.org/ [2] https://deepmind.google/models/synthid/ [3] https://github.com/google-deepmind/synthid-text

There is a repo: https://github.com/google-deepmind/synthid-text

Google's SynthID is embedded into the content itself. Google open sourced their SynthID for text. Repo: https://github.com/google-deepmind/synthid-text Paper: https://www.nature.com/articles/s41586-024-08025-4 With images and video, it's less clear exactly what they're doing, but it's watermarking on the pixel leve. From one of their blog posts: Videos are composed of individual frames or still images. So we developed a watermarking technique inspired by our SynthID for image tool. This technique embeds a watermark directly into the pixels of every video frame, making it imperceptible to the human eye, but detectable for identification. https://deepmind.google/discover/blog/watermarking-ai-genera... Elevenlab's audio watermarking is trivial to shake off with compression, but google claims that synthid is resilient to such manipulation.

Some links from the original article Watermarking AI-generated text and video with SynthID https://deepmind.google/discover/blog/watermarking-ai-genera... synthid-text https://github.com/google-deepmind/synthid-text