AS, in short, it is easy to measure pulse or sleep somehow. It is hard to measure it well consistently (pulse when someone is running, biking or weightlifting, sleep when people sleep with others, move, etc - or lay sleepless).

https://news.ycombinator.com/item?id=37469920 (from the same org)

I mean, heart rate? Do we have a giant network that can tell where everybody is and whether they are having a strong emotional response to anything?

- Whether train/test splits were participant-wise to avoid data leakage.

- How the system performs at elevated or highly variable heart rates.

- Results from "placebo" or empty-room baselines to rule out false positives, typically done with bags of rice/water (used to simulate mass).

Impressive from a technical standpoint, but super scary from a privacy standpoint. Surely this must allow them to detect and differentiate between plosives, which is probably enough to infer what's being talked about.

https://www.nature.com/articles/s41598-024-77683-1

That kind of module can be really cheap

https://www.waveshare.com/hmmd-mmwave-sensor.htm

and is starting to replace Passive IR sensors

https://en.wikipedia.org/wiki/Passive_infrared_sensor

I can imagine one of these on the ceiling above your bed being an ideal sleep monitoring system.

Google: serves you Aspirin ads when they notice you are having a heart attack