[theclaw]

Given the behaviour demonstrated (register fingerprint, everything works as expected, add screen protector, any fingerprint unlocks), I'd imagine that they're not dealing with every possible result the fingerprint reader can produce. For example, they might be assuming the result could be either a pass or a fail, but in fact there are three states: valid fingerprint and matched, valid fingerprint and not matched, and not a valid fingerprint.

Perhaps the fingerprint reader returns a result of 0 for "valid print and not matched", result 1 for "valid print and matched", and result -1 for "not valid print". If the phone vendor code simply tests for a non-zero result code and treats this as "matched", then the situation described would occur.

It would be interesting to test with other smooth objects that might activate the reader.