Researchers track keystrokes using smartwatches
A US university has developed a method for using the motion sensors on smartwatches to guess what the user is typing, in a discovery the team says has major privacy implications.
University of Illinois researchers used a homegrown app on a Samsung Gear Live smartwatch to capture motion data from the in-built accelerometer and gyroscope and use it to detect typing activity.
The researchers believe that attackers could potentially develop a similar app, disguise it as a fitness tracker or a similar legitimate service, and upload it to mobile app stores.
But the discovery isn't likely to keep security researchers up at night just yet — the app was able to accurately detect keystrokes, but only on the hand wearing the watch.
The prototype system as yet also cannot detect special characters or the keystrokes from users with different typing patterns. The use of the space bar also adds complications.
The researchers nevertheless believe that the vulnerability does represent a potential significant security problem for any wearable device that uses motion sensors.
According to the University of Illinois, this problem could be addressed by lowering the sample rate of sensors in smartwatches to prevent the accurate measurement of typing activity, the researchers said.
Why the Five Eyes AI warning should change the Board's cyber agenda
If you don't test your identity recovery in a real-life crisis scenario, you are...
The next cyber threat has arrived and identity security is key for survival
Frontier AI is poised to reshape cybercrime by making sophisticated attacks faster, cheaper and...
Faking a trusted brand is easier than ever, defending against impersonation isn’t
Organisations need to defend their brands with renewed vigour as AI-driven...
