I have a mental test that I always do when trying to evaluate the privacy protection guarantee offered by an online service, it goes like this: imagine if a mal

I have a mental test that I always do when trying to evaluate the privacy protection guarantee offered by an online service, it goes like this: imagine if a malicious actor is holding the top most owner of the service at gun point, and demanding for my user data, will the owner risk their own life to obey the privacy agreement? Probably not. Of course, the mental test is very extreme, but it proofs that the guarantees offered will fail and be violated when some threshold is exceed. It's not a sign and done deal.