Beware! This AI can steal your passwords by 'listening' to you type

A recent study conducted by researchers reveals that malicious individuals could potentially exploit artificial intelligence (AI) tools to achieve near-perfect precision in stealing user passwords. The study, carried out by Cornell University based in the United States, highlights that when the AI software was activated on a nearby smartphone, it was able to accurately replicate the entered password with an astonishing accuracy rate of 95 per cent.

The research involved a team of computer scientists from the United Kingdom who trained an AI model to identify the sounds produced by keystrokes on a 2021 version of a MacBook Pro. During a Zoom video conference, the AI tool exhibited remarkable accuracy in "listening" to keystrokes picked up by the laptop's microphone.

The researchers noted that the AI program was able to reproduce the keystrokes with an impressive accuracy rate of 93 per cent, setting a new benchmark for this method of attack. Additionally, the researchers emphasised a concerning aspect that many users are not aware of malicious actors could potentially monitor their typing to exploit vulnerabilities and gain unauthorised access to accounts. This form of cyberattack is known as an "acoustic side-channel attack."

A statement by Zoom read: “Zoom takes the privacy and security of our users seriously. In addition to the mitigation techniques suggested by the researchers, Zoom users can also configure our background noise suppression feature to a higher setting, mute their microphone by default when joining a meeting, and mute their microphone when typing during a meeting to help keep their information more secure.”

Also Read Hollywood vs AI: Why famous actors including Oppenheimer, Barbie cast are on strike

What is an acoustic side-channel attack?

An acoustic side-channel attack is a type of cyberattack that leverages unintended sound emissions or vibrations produced by a computing device to gather sensitive information. Side-channel attacks are a class of attacks that exploit information leaked during the execution of a cryptographic algorithm, such as timing, power consumption, electromagnetic radiation, and in this case, acoustic signals.

In the context of an acoustic side-channel attack, the attacker uses specialised tools or techniques to capture the acoustic emissions produced by a device during its operation. This could include sounds generated by keystrokes on a keyboard, clicks on a mouse, or even the subtle sounds produced by internal components as they process data. These emissions can carry valuable information about the device's operations, such as the timing and sequence of keystrokes or other user inputs.

By analysing the captured acoustic signals, an attacker can potentially infer sensitive information, such as passwords, PINs, or other confidential data, that is being entered by a user on the targeted device. This type of attack is particularly concerning because users often do not realise that these acoustic emissions can be exploited to compromise their security.

The study underscored the widespread nature of keyboard acoustic emissions, which not only provide an easily accessible avenue for attacks but also lead individuals to underestimate the risk associated with these emissions and consequently fail to take precautions. For instance, individuals tend to shield their screens while typing passwords but pay little attention to concealing the sound produced by their keyboards.

To evaluate the precision of the AI program, the researchers meticulously pressed each of the laptop's keys 25 times, varying factors such as pressure and finger position for each press. The AI software had the capability to "listen" for distinctive characteristics of each key press, such as unique sound wavelengths. Positioned at a distance of 17 centimetres from the keyboard, the smartphone used for testing was an iPhone 13 mini.

Also Read

Battle of the billionaires: Elon Musk vs Mark Zuckerberg cage match could make over $1 billion

Google appeals to Supreme Court to quash antitrust directives on Android in India