New WiFi tech can accurately identify individuals without devices

Researchers at the Karlsruhe Institute of Technology (KIT) have developed a novel WiFi-based technology capable of accurately identifying individuals without requiring them to carry any connected devices. By analyzing the interactions of WiFi signals emitted by standard devices in the environment, the system can generate detailed representations of a person’s presence, posture, and movement, comparable to images captured by cameras but constructed entirely from radio waves. This method works even if the person’s devices are turned off, as it leverages signals from other active WiFi devices nearby, raising significant privacy concerns. The technology exploits unencrypted beamforming feedback information (BFI) regularly exchanged within WiFi networks, enabling the creation of multi-angle images that can be processed by machine learning models to identify individuals within seconds. Unlike previous approaches requiring specialized hardware, this technique uses only standard WiFi devices already present in homes, offices, and public spaces, making it easier to deploy and potentially widespread. Experts warn that this capability could allow surveillance by public authorities or companies without individuals