Hackers could steal passwords by listening to keystrokes using AI: Study
AI can steal passwords from keystroke sounds recorded over Zoom with up to 93% accuracy, per a new study. The accuracy rate ratcheted up to 95% when keystrokes were recorded using an iPhone 13 mini.
FollowAn AI tool could decipher text — including passwords — from keystroke sounds recorded over Zoom and be right over nine times out of ten, a group of researchers said in a paper published on August 3. According to the study, AI is able to recognise the particular keys being pushed simply listening to the typing noises.
Researchers stressed that the risk of sound-based assaults has grown with the growing use of video conferencing services like Zoom and the ubiquitous availability of devices equipped with microphones in a report released on August 3. The researchers' AI programme demonstrated the ability to reliably interpret text from keyboard noises, including passwords, with a precision level above 90%.
Also Read |Â Google rolls out new AI search on Chrome to summarise web pages
The study, which includes researchers from the universities of Durham, Surrey, and Royal Holloway, demonstrated how microphones can recognise certain typing patterns. According to this information, anyone who use computers in public places may be at danger of having their typing recorded and later decrypted.
After attempting to train an AI model, the study team came to this result. They achieved this by pressing each of the 36 keys on a MacBook Pro 25 times while recording the accompanying sounds. Subsequently, they fed this sound data into the AI model, enabling it to accurately recognize the unique pattern associated with each key.Â
Also Read |Â Apple AirPods likely to be manufactured at Foxconn's Hyderabad factory: Report
For their experiment, the team used a 2021 16-inch MacBook Pro, highlighting its consistent keyboard design with other recent MacBook models.
With the use of the Zoom video conferencing programme, the researchers' AI model demonstrated an impressive accuracy rate of 93% when deciphering keystrokes from recorded MacBook typing noises. Additionally, when the keystrokes were recorded using an iPhone 13 mini, the accuracy rate climbed to 95%.
