A post just in case nobody knows about it. This isn't a conspiracy post, it can be done, the video link is there by Cyb3rMaddy to explain it more thoroughly. It isn't new tech, it has been around for more than 10 years. It is just crazy to think that somebody can somehow tap into it, maybe your ISP. Who knows how this is being used today. Since there are no cameras on routers, it is unregulated. Your ISP can theoretically use your home Wi-Fi router to track your movements. By analyzing how your body and walls disrupt and reflect wireless signals a technology known as Wi-Fi sensing, AI models can map exactly where you are, what you're doing, and even your posture. This is wild to think about, I am guessing regular people don't know this, but it can happen. Since this can be done, somebody can use it against you. It can even detect human breathing rhythm and heart rate. Sitting at home and unaware. Your cell phone can be used as an Wi-Fi access point. Another example is your smart fridge that has Wi-Fi, it does send back data to somewhere anyway. Like when you run out of food or something. I would never get one of those, personally. I totally understand those people that just want to move to a cabin in the woods somewhere.
https://www.sciencedaily.com/releases/2026/05/260522023127.htm
Researchers say these signal reflections can effectively create multiple "views" of a person, allowing AI systems to learn and recognize individual identities.
After the machine learning model has been trained, identifying a person reportedly takes only a few seconds.
Original article: https://www.kit.edu/kit/english/pi_2025_069_the-spy-who-came-in-from-the-wifi-beware-of-radio-network-surveillance.php
It works by exploiting the communication of legitimate users of the WLAN, whose devices are connected to the WiFi network. These regularly send feedback signals within the network, also called beamforming feedback information (BFI), to the router – in unencrypted form so that it is readable by anybody in range. This creates images from different perspectives that can serve to identify the respective persons. Once the underlying machine-learning model has been trained, the identification only takes a few seconds.
Paper: https://publikationen.bibliothek.kit.edu/1000185756/168100988
Edit: citation from original article