Sadly, I think the only way to protect against this is with another AI whose job it is to recognize what data is appropriate to reveal and what is private - bas

Sadly, I think the only way to protect against this is with another AI whose job it is to recognize what data is appropriate to reveal and what is private - basically what humans do. But, even then it will probably still be susceptible to tricks. Of course the ideal thing is just to not include it in the training data but I think we know how much effort that would take when the training data is basically the entire internet. I wonder if as AI systems become more efficient and they learn to "forget" information which isn't important and generalize more, that this will become less of an issue.