4 ms·

Anthropic to start embedding invisible watermarks

• TruffleLabs 2mo ago
  Archive.is link https://archive.is/8n95Q https://archive.is/8n95Q
• TruffleLabs 2mo ago
  Not sure how text (ascii) can be watermarked ...
  • Avicebron 2mo ago
    https://arxiv.org/abs/2301.10226 https://arxiv.org/abs/2301.10226
    • FabHK 2mo ago
      Just to be clear, as I understand it, that's not the method.
  • ashu1461 2mo ago
    Ascii can't but I guess most of the text on the internet is unicode which easily can be.
  • perfmode 2mo ago
    biased word choice
• russfink 2mo ago
  Does anyone have a link to the method? Does it need Unicode or is it ASCII spacing? Word frequency bias? Can a human inadvertently trigger a detector for it?
  • FabHK 2mo ago
    ICYMI: https://news.ycombinator.com/item?id=49303350 https://news.ycombinator.com/item?id=49303350 Basic idea is that when picking a next token from the probability distribution from the model, they use a pseudorandom generator (ie deterministic) whose signature they than can pick up.