> The sensors of the air purifier can be used for spying

To be able to... know if your target's house has a lot of pollutants? Is particularly warm? There is practically no useful information that can't be gleamed by just looking through their windows, blinds and all.

> and the device could also be used as a hopping point for exploiting other devices in your home.

It's not connected to your home network, that's the whole reason for the hotspot existing. How, exactly, could it be used as a hopping off point, except to other devices with hotspots that... can just be exploited in the first place.

You should do some tests with air quality, humidity and temperature sensors and plot them on a graph to see how they can be good indicators of room occupancy and even the level of physical activity. Not all spaces have easy visual access.

There is also a difference between having seconds to exploit a known-vulnerable device as you drive by or being able to take your time from a local device with a Wi-Fi antenna that you now control.

Obviously a wrench may work better than these possible access points, but the point is that these vulnerabilities should not exist, and that they don't need to exist for the device to serve its function.

You're lacking in imagination, and maybe the conceptual idea of "sensor fusion". Multiple seemingly innocuous data streams in isolation can be combined to create sensors you wouldn't have imagined

Do you understand what data is available in a smart air purifier?

Please, explain exactly what sensor fusion would get you actionable data out of the PM2.5 sensor and "gas sensor" in a Philips smart air purifier.

At a guess; if able to monitor over a period of time (e.g. pick up data from a parked car), a potential burglar can see when there is activity and figure what times of the day house occupants are normally at home.

> that can't be gleamed by just looking through their windows, blinds and all.

I mean, sure, but who is going to do that when they can... look to see when people are home.

More subtle; the burglar could just park up and go off for a few hours and gather the data they need - no need for a suspicious camera pointing out of the car to monitor patterns.

If the burglar only takes a 30 second look before breaking in, residents could be home but away from a window, with this the burglar can more confidently know when is a good time to break in, without exposing themselves to the same risk that looking around the house brings.

Maybe increased CO2 on Tuesday afternoons will tip off that the wife is cheating?

It's not even far fetched, smart watches reporting physical actively at unexpected hours have revealed infidelity in the past.

If the sensors don't detect your farts for a while you're probably not at home so the burglars can come in

Just a mere few years ago you wouldn’t believe WiFi access point can be used as a sonar to literally scan the area like a low-def camera in real time.

Stuxnet also sounded like a completely made up scenario.

As someone said, you lack imagination. And that’s OK, but you’re also being quite arrogant, too.

> There is practically no useful information that can't be gleamed by just looking through their windows, blinds and all.

I have plenty of imagination. I also am practical and realize how illogical the argument of “sensor fusion” is to do something you can just use your eyes for.

Nobody is going to go out of their way to do this when KISS methods already exist and y’all don’t seem to understand practicality if you don’t see that.