Researchers at Pennsylvania State University have developed a groundbreaking technology that can eavesdrop on phone conversations from a distance of up to three meters, utilizing millimeter-wave radar and artificial intelligence. The system achieves an accuracy of about 60% by analyzing the minute vibrations generated by a phone's receiver during a call. This was reported by Interesting Engineering.

The study is based on the adaptation of the open-source speech recognition model Whisper, retrained on radar data, which significantly improved speech recognition quality, even when the data is noisy and low-quality. The radar was placed approximately three meters away from the phone, and the data was processed by AI capable of recognizing up to 10,000 words.

Although the system's accuracy isn't perfect, even partial recognition of key words can have serious implications for privacy. The researchers liken their technology to lip-reading; it may not capture every word but helps understand the essence of conversations, especially when considering context.

It's crucial to note that this technology raises concerns about potential privacy threats, as malicious actors could use it for remote eavesdropping on personal conversations without users' knowledge. The researchers emphasize that their work aims to raise public awareness about such risks and contribute to the development of protective measures against this type of spying.

The research is supported by the National Science Foundation, and the team plans to continue working on safeguarding personal information from emerging technological threats.