Your fingerprints can be recreated from the sounds made when you swipe on a touchscreen.

turkeydance

turkeydance

Well-Known Member
Life Member
Supporting Member
Joined
Jul 17, 2017
Messages
4,507
Location
nc
Rating - 100%
4   0   0
quote:
“finger-swiping friction sounds can be captured by attackers online with a high possibility.”
The source of the finger-swiping sounds can be popular apps like Discord, Skype, WeChat, FaceTime, etc.
Any chatty app where users carelessly perform swiping actions on the screen while the device mic is live.
Hence the side-channel attack name – PrintListener.

www.tomshardware.com

Your fingerprints can be recreated from the sounds made when you swipe on a touchscreen — Chinese and US researchers show new side channel can reproduce fingerprints to enable attacks

Researchers claim they can successfully attack up to 27.9% of partial fingerprints.
www.tomshardware.com www.tomshardware.com
 
Just another reason to chuckle at all these people streaming non stop. Giving away their voiceprint to anyone willing to have AI listen.

Ill pass
 
AI detects A-holes!

"......can be reproduced from the sound of TP sliding across...."

note to self, turn off mike in the john.
 
The moment you realize you used your voice mail greeting as the voice command to unlock your office safe.
1000001781.jpg
 
Last edited:
I wonder what kind of prints they come up with when you're swiping with something that isn't a finger...
 
GymB said:
I’ll call BS on this, simply don’t believe it.
Click to expand...
Well, at a previous job the task was to determine the EFI risk in relation to username and passwords. That's far as I can go. So yes, repeated attempts and RFI analysis can determine those.

As the "swipe" I'm not sure as by the time EM is in play any physical attributes are out. There is no finger noise, only (redacted) and (redacted). If you want to pay I can play more...
But the .gov will be involved. I'm not sure i want to go back into that.
 
Last edited:
rdinatal said:
Well, at a previous job the task was to determine the EFI risk in relation to username and passwords. That's far as I can go. So yes, repeated attempts and RFI analysis can determine those.

As the "swipe" I'm not sure as by the time EM is in play any physical attributes are out. There is no finger noise, only (redacted) and (redacted). If you want to pay I can play more...
But the .gov will be involved. I'm not sure i want to go back into that.
Click to expand...
Lol, I’m not saying that someone couldn’t be identified. Things like cadence when typing are great for that, I assume that there are a number of attributes that make a swipe fairly unique. But, the capability of the phone to capture the sound of the swipe as described, distinguish that from background noise picked up by the microphone and then not just distinguish one person from another but create an image of the users fingerprint, it just isn’t happening.
 
  • Like
Reactions: HMP
well, in the report, they admit it is not that good Right Now....

"...successfully attack up to 27.9%..."

but they are working on it.
 
Some people simply have no clue what AI is capable of, and it's still just a baby. Those that aren't in the industry scoff.
 
You must log in or register to reply here.
Back
Top Bottom