318
LLMs can unmask pseudonymous users at scale with surprising accuracy (arstechnica.com)
submitted 8 months ago by to c/technology@lemmy.world
108 comments fedilink hide all child comments
you are viewing a single comment's thread
view the rest of the comments
[–] 40 points 8 months ago (3 children)

And it will falsely identify people at even greater scale, because it is an imprecise and buggy tool.

  • permalink
  • fedilink
  • source
  • hideshow 3 child comments
    •
  • [–] 2 points 8 months ago

    The ones doing the identifying will argue that they are right until they are blue in the face because that's how we do things now. By lying harder until the ones you're trying to convince believe you or give up because it's not worth it to keep arguing. I've been accused of being other people or using AI to write, and those people argued or harassed me until I blocked them. I can probably expect more of this going forward.

  • permalink
  • fedilink
  • source
  • parent