299
How AI text watermarking works: a visual guide (declaude.org)
submitted 3 weeks ago by to c/technology@lemmy.world
105 comments fedilink hide all child comments
you are viewing a single comment's thread
view the rest of the comments
[–] 19 points 3 weeks ago (2 children)

It works for them not scraping their own slop back into training data. I assume that is actually the real purpose of the system. They don't want to share the key with the public. But they probably will with other llm companies in exchange for theirs.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 4 child comments
    •
  • [–] 8 points 3 weeks ago (1 child)

    Hmm, yes. I just thought about outside usage, somehow haven't thought about it as an inner LLM maintenance tool.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] 1 point 3 weeks ago (2 children)

    The article seemed to claim that some models allow outsiders to submit text for detection if I read it right. That seems like a decent way of doing it. If you "open source" the raw data, it means people can do things to try and get around it - same reason most websites don't reveal their anti-spam techniques.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 4 child comments
    •
  • [–] 2 points 3 weeks ago (1 child)

    That's still not very useful unless the test is for all AI models otherwise you have to manually submit the text to every AI companies detection system.

    I see an opportunity for an AI test aggregator company that checks all of them for you at once.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] 2 points 3 weeks ago

    Yeah, but you'll have to submit to every known provider and hope the user used one of the ones that provide this checking service, and didn't do something like ask a local model to just randomize synonyms in a text.

  • permalink
  • fedilink
  • source
  • parent
  • [–] 0 points 3 weeks ago*

    That would only work for their own slop though. Anthropic cannot recognize Google's watermark, only theirs.

    I assumed the goal might be so they could check whether other models have been trained on their output. Like anthropic using that as a "proof" when they start whining again about Chinese "distillation attacks".

    Of course, since they're the only ones able to check their watermark, it would be rather shit as evidence anyway. "We've run the numbers, and we know you can't, but trust us, this chatbot is totally copying ours!"

  • permalink
  • fedilink
  • source
  • parent