Spymarks just seem like another word for <a href="https://en.wikipedia.org/wiki/Steganography" rel="nofollow">https://en.wikipedia.org/wiki/Steganography. On that note, one way we can prevent it is to assert that all our content is byte-for-byte identical with the last known trusted stage of what we have produced (for example: a camera we are certain does not watermark, an image editor we are certain doesn't watermark, an image compressor we are certain can't watermark, etc). One vector that I am particularly concerned about is social media. Most images and videos uploaded to most social media is re-compressed by the target platform. This is a door to tracking that is far too easy for social media platforms to open. They might rationalize it (if discovered/announced) by saying that our memes won't be reposted, images or work stolen, etc... but honestly I'd rather my work be stolen than tracking information inserted in there. Oh, we also have stuff which is way more secure, like time-stamped cryptographic signatures.
I actually wrote a library to do stego with LLM outputs last year and it turned out to be an almost exact implementation of the Anthropic watermark algo.
Repo here <a href="https://github.com/sutt/innocuous" rel="nofollow">https://github.com/sutt/innocuous. It works with last year's llama.cpp. Check out the "Use Cases" and "How it works" sections in the readme if you're interested.
Retro_Dev · · focus · HN ↗
stillsut · · focus · HN ↗
Repo here <a href="https://github.com/sutt/innocuous" rel="nofollow">https://github.com/sutt/innocuous. It works with last year's llama.cpp. Check out the "Use Cases" and "How it works" sections in the readme if you're interested.