63
New AI attack can reconstruct typed text from keyboard sounds (cyberinsider.com)
submitted 3 months ago by to c/technology@lemmy.world
31 comments fedilink hide all child comments
you are viewing a single comment's thread
view the rest of the comments
[–] 77 points 3 months ago (13 children)

This doesn't need AI, it has been done for a long time

  • permalink
  • fedilink
  • source
  • hideshow 13 child comments
    •
  • [–] 6 points 3 months ago* (5 children)

    I don't know how the existing solutions work, but the article mentions it being possible without training on the specific keyboard first and with 99% accuracy. And that would be frightening, especially combined with that sentence:

    The researchers describe three realistic attack scenarios, where an attacker could:

    • ...
    • Recover keystroke sounds transmitted as background noise during online meetings

    How would I write shit about the idiots presenting in the meeting without them knowing?

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 5 child comments
    •
  • [–] 7 points 3 months ago* (4 children)

    Luckily software like Teams has noise cancellation, so keyboard sounds generally get filtered out.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 4 child comments
    •
  • [–] 1 point 3 months ago (2 children)

    The researchers further tested the attack against audio transmitted over Google Meet, Microsoft Teams, and Zoom. With noise suppression disabled, keystroke sounds carried through meeting audio were often sufficient to reconstruct typed text with high accuracy after a few hundred keystrokes, although results varied depending on the conferencing platform and laptop model. The paper notes that some users may still be exposed because advanced noise cancellation features are not always available or enabled.

    Not so fast...

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] 2 points 3 months ago (2 children)

    And now it's easier and more accessible.

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

    Yeah, I don't really get these "this has always been possible" comments whenever it is about AI. Yes you could do this before, yes you could manipulate or even completely and convincingly make up photos before AI.

    The point is that now is that it actually took some time, skill and effort to do that before, but now this has basically been made easily available to everyone, taking no effort or time. So that now any malicious moron out there who wants to do something destructive and evil are actually capable of doing it on the spur of the moment.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 1 child comment
    •
  • [–] -4 points 3 months ago

    Nono no no no! AI is dumb and ineffective 🙄

    This place is just progressive Reddit 2.0. It's a time waster. Expecting and nuance or useful conversation around anything that doesn't fit the communities world view will result in the same shit as Reddit.

  • permalink
  • fedilink
  • source
  • parent