AI Can Now Record Your Passwords Just By Listening

Remember the keypad phones we had once upon a time? We could tell what key we were pressing just by listening to them. The level of accuracy that general users had reached in identifying the keys was quite uncanny. Artificial Intelligence now adopts the same behavior, as it can now understand what you’re typing just by listening to the tapping sound of the keyboard or on your phone screen.

This technological capability is a big tool for cybercriminals to capture sensitive information like passwords, credit card credentials, etc. This is not an advancement achieved by an unknown AI enthusiast but is part of research published in the IEEE European Symposium on Security and Privacy Workshops. The lead researcher Toreini and his colleagues have used machine learning to build an AI-powered algorithm that enables a system to identify what keys are being pressed on a device by listening to it with 90% accuracy. This development comes as a second step of the Enigma Cipher device

The research, published as part of the IEEE European Symposium on Security and Privacy Workshops, reveals how Dr. Ehsan Toreini and colleagues used machine learning algorithms to create a system able to identify which keys were being pressed on a laptop based on sound – an approach that researchers deployed on the Enigma cipher device in recent years. 

The Danger Lies In Abundance Of Ears

I can only see the accuracy of such models, and such attacks, increasing!

 Toreini, co-author of the study at the University of Surrey

As technology advances, it becomes a closer part of our daily lives too. Just look around yourself, and you’ll find at least three devices connected to the internet with an in-built microphone. Whether it is your phone, laptop, or home Alexa device, the capability to listen to you in the most crowded and isolated places is almost 100%. 

Nearly 12.92 million people use Zoom to interact with their teams every month. Having the AI tool installed in one of the devices on the call can do the job of listening to every person typing with up to 90% accuracy. All they need to do is have their mics switched on. 
The researchers have taught their AI tool to recognize the acoustic signals associated with each key. While it is unclear how the tool does it, it poses a strong possibility of identifying the position of the key pressed to know which one it was.

Written By

Manish Content Strategist

With a mixture of literature, cinema, and photography, Manish is mostly traveling. When he is not, he is probably writing another tech news for you!

Follow