I'm wondering how long before some enterprising thieves figure out how to lift your fingerprint off your device and feed it into the fingerprint reader?
For the current type of sensor, something like that was first done back around 2001.
Interesting question about ultrasonic though... is it just as easy to fool with the right hobby tools? The claim is that it provides a more detailed 3D image, a deeper one, and can more easily detect "liveness". Yet in the past, even sensors with such detection have been fooled by fake pulses, or even the pulse of the wearer of the fake print.
I wonder what technology Apple uses for Touch ID, if it can be describes in a word like ultrasonic?
The current Touch ID is an RF capacitive type. Those have been around since at least the late 1990s.
In Touch ID, it's a high resolution (500 DPI) array of tiny flat antennas. Similar to a touch panel, but much denser.
An RF signal is transmitted from the outer metal ring into your finger, and the signal levels (corresponding to your fingerprint ridges and valleys) are detected on each antenna "pixel" in the array, basically building a 3D (2D array of different values) image of your print.
Btw, the problem with a wet sensor, is that allows the RF signal to short circuit straight from the ring across the flat outside, making it more difficult to read the internal signals from the finger itself. Ultrasonic does not have that electrical problem of course.