Could AI be listening in on you? It seems so. Researchers from renowned universities in the UK have created a groundbreaking AI system capable of eavesdropping on keyboards to harvest potentially sensitive data. This algorithm, detailed in a recent paper, achieved an impressive 93-95% accuracy in detecting pressed keys solely based on audio recordings from a MacBook Pro keyboard.
TL;DR:
- AI system capable of eavesdropping on keyboards, detecting pressed keys with 93-95% accuracy based on audio recordings from a MacBook Pro keyboard.
- Microphones in consumer devices make acoustic side-channel attacks possible, posing a risk to data security.
- Protect against AI eavesdropping by confusing the model with different typing styles or using touch-screen keyboards. Stay vigilant about emerging AI threats to safeguard data and privacy.
The AI Listening Study
The study sheds light on the widespread use of microphones in consumer devices that could be exploited for acoustic side-channel attacks. The consumer devices are phones, laptops, tablets, etc. Therefore, the possible exploitation poses a risk to data security. Unlike previous approaches to laptop keystroke detection via audio, this AI-based method has incredible precision.
According to the researchers, their AI model is much more accurate than other hardware-based methods. Other methods often face limitations due to distance and bandwidth constraints. However, as microphones are now embedded in common consumer gadgets, the acoustics of typing have become increasingly vulnerable and accessible.
So how does this cutting-edge audio algorithm work? The researchers recorded audio samples of typing on a MacBook Pro. They pressed each key 25 times. This then allowed the AI to discern tiny variations in the sound emitted by each key.
Subsequently, the audio recordings were transformed into spectrograms. Spectrograms provide visual representations of sound frequencies over time. Moreover, the AI was trained on these spectrograms, learning to correlate distinct patterns with different keystrokes.
So, by applying this training across thousands of audio segments, the algorithm acquires an incredible understanding of the acoustic fingerprints unique to each key. Then, once trained on a specific keyboard model, the AI can accurately analyze new audio recordings. With the recordings, it can predict keystrokes – and know what you are typing. The AI with MacBook Pro keyboard training achieved an impressive precision rate of 93-95%. Even when the sound came from Zoom call recordings, the AI’s performance only dropped slightly.
How Can I Protect Myself Against AI?
There is no need to panic (yet). The AI system needs calibration for specific keyboard models and audio environments. So, in order for a hacker to use AI to gain data, they would need to have access to the necessary training data. Still, if you want to stay extra safe, there are a few actions that you can take. Firstly, confuse the AI model by pressing keys differently. This reduces its accuracy by 40%. Secondly, use speakers to play sounds. You could even swap your keyboard for a touch-screen keyboard.
As AI’s capabilities grow, so does the worry of emerging threats. Therefore, it’s important to stay up to date on all the latest AI advancements and secure your data and privacy.
