Exactly. All of these stories using WiFi to detect things with high accuracy are just extreme machine learning demos.

Given a tightly controlled environment and enough training data, you can use a lot of things as sensors.

These techniques are not useful for general purpose sensing, though. The WiFi router in your home isn't useful for this.

WiFi AP's already do a lot of tracking and measurement just to improve signal fidelity and effective throughput. Why wouldn't those same techniques be useful for more general object tracking? Of course using a single AP to attempt to track movement in real-time is unlikely to have great results, but with several APs and enough compute triangulation should improve results.

today's tech demos are tomorrow's everyday