First, it was our webcams and microphones, but now our keyboards aren’t safe either. People using laptops are now in danger of having their confidential data, such as private messages, passwords, and credit card details, compromised solely through their typing activity.
A recent research paper authored by a consortium of British university scholars underscores that artificial intelligence possesses the capability to discern keystrokes based solely on auditory cues, achieving an accuracy rate of 95%. These attacks are only expected to become more sophisticated given how fast AI is developing these days.
How it Works
The research paper delves into the realm of what it terms “acoustic side channel attacks,” wherein a malicious entity employs an auxiliary device, like a cell phone positioned adjacent to a laptop, or an un-muted microphone on a video conferencing platform such as Zoom, to capture the auditory input of typing sounds.
Subsequently, this captured sound is processed through a deep-learning artificial intelligence model that has been trained to discern the distinct auditory patterns of individual keystrokes, ultimately revealing the content of the typed text.
Using this method, the researchers adeptly identified keystrokes executed on a MacBook Pro by solely analyzing the associated sound via a nearby mobile phone, attaining an impressive success rate of 95%. Similarly, through the analysis of a recorded Zoom call, the identification accuracy remained highly alarming at 93%.
The paper says:
With the recent developments in both the performance of (and access to) both microphones and DL models, the feasibility of an acoustic attack on keyboards begins to look likely.
How to Avoid Such Attacks
One way of avoiding such attacks would be to use stronger passwords with multiple cases, so the AI has trouble recognizing all the keystrokes combined. Also, passwords with full words will be easier to crack compared to just a jumble of numbers, letters, and other characters.
You can also add other simple security measures such as two-factor authentication as well as biometric verification. But the paper warns that as AI evolves, it may be able to tackle various other security prompts in the future.
Via: Fortune
Follow ProPakistani on Google News & scroll through your favourite content faster!
Support independent journalism
If you want to join us in our mission to share independent, global journalism to the world, we’d love to have you on our side. If you can, please support us on a monthly basis. It takes less than a minute to set up, and you can rest assured that you’re making a big impact every single month in support of open, independent journalism. Thank you.
guess people who don’t have mechanical keyboards have nothing to worry about.
seems baseless for these as 2FA and mouse are also being used in captcha.