There is a phrase that sounds different now: “WiFi is everywhere.” Yes, we know. At home, at the cafe where you write, at the airport, at the office, at the restaurant on the corner. But what you probably didn't know is that all those invisible signals floating around you can now be used to recognize you, track you, and even identify you with an accuracy that borders on 100%. And the worst thing, without you having any device on you.

A team of researchers from the Karlsruhe Institute of Technology (KIT) in Germany published a study that shook the global technology community. They demonstrated that conventional WiFi routers, the same ones you have at home or in any public space, are capable of identifying people with an accuracy of 99.5%. They don't need cameras. They do not need you to be connected to the network. They don't need you to carry your cell phone with you. They just need WiFi to be active where you are.

How WiFi waves become an “invisible camera”

To understand how this works, we first have to talk about something you've probably never heard of before: BFI signals, or Beamforming Feedback Information. Modern routers compatible with WiFi 5 and later use this mechanism to optimize their connection to nearby devices, basically to better “point” the signal where it is needed.

The problem is that these signals travel without encryption, which means that anyone within the same physical space can capture them without needing to know the network password or touch the router. And that's where black magic comes in, or rather, artificial intelligence.

What the KIT researchers discovered is that when a person walks or moves within the area covered by a wireless network, their body alters the radio waves around them in a unique way. This alteration generates a specific pattern that depends on how that person walks, the shape of their body, their movements. A pattern that, after being processed by artificial intelligence models trained with thousands of examples, allows one individual to be distinguished from another with chilling accuracy.

Professor Thorsten Strufe from KIT's Information Security and Reliability Institute put it simply: “By observing the propagation of radio waves, we can create an image of the environment and the people present.” It's like a camera, but instead of using light, it uses radio waves that are already there, floating in the environment.

The experiments were carried out with 197 participants in controlled environments and the results left no room for doubt: the system managed to identify each person with an accuracy of 99.5%. And the most disturbing thing of all is that the hardware necessary to achieve this is neither sophisticated nor expensive: a laptop, a Raspberry Pi-type device and a WiFi card are enough.

The privacy you thought you had no longer exists like before

Julian Todt, a doctoral researcher at KIT, was quite direct with the implications: “If you regularly walk past a coffee shop that operates a WiFi network, they could identify you there without you noticing and recognize you later.” That sentence is worth reading twice.

It's not just about someone knowing that “there is a person in this room.” The system can recognize you, specifically you, every time you pass through a space covered by WiFi. And since the planet is literally covered in active wireless networks, the scenario this opens up is enormous.

It doesn't matter if your phone is turned off, if you left your laptop at home or if you are not connected to any network. As long as there are other people or devices nearby communicating via WiFi, the signals from the environment are enough to detect you and differentiate you from others. That turns each active router into what the researchers themselves described as a “silent observation device.”

The team also pointed out something that should worry more than one: in authoritarian or mass surveillance contexts, such technology could be used to monitor political opponents, journalists or activists without the need for any visible camera, without physical controls and without leaving obvious traces of surveillance.

The doors this opens and the risks that no one wants to see

It would be dishonest to say that this technology has only obscure uses. The same ability to track people without their consent could also transform entire industries.

Imagine a home security system that detects intruders without cameras, or medical devices that monitor the movements of elderly patients within their homes without the need for wearables or additional sensors. MIT researchers had explored something similar before, showing that WiFi waves can detect not only the presence of people but also their body postures and health states.

In the world of retail, the possibilities are just as disruptive and controversial: identifying frequent customers when entering a store, studying movement patterns within physical spaces or personalizing experiences in real time without the user having to do anything at all.

But the consensus among the research community is that the urgency is on the side of protection, not exploitation. The KIT team has already explicitly called for future wireless networking standards to include more robust protections for BFI data, precisely because that is the gateway that makes all of this possible.

The paradox is uncomfortable: the same infrastructure that connected us to the digital world is becoming the most discreet surveillance tool in existence. You don't need a spy satellite or a team of undercover agents. Just the router you have next to you right now is enough.