> Of course, if you value the privacy of these data a lot, and if you believe that by allowing them to train on them it is actually going to risk exposing priva

> Of course, if you value the privacy of these data a lot, and if you believe that by allowing them to train on them it is actually going to risk exposing private info, the story changes. But then you have an option to say stop. It's up to you to measure how much you value "getting a better product" vs "estimated risk of exposing some information considered private". Some will err on one side, some on the other. The problem with this reasoning, at least from what I am understanding is that you don't really know when/where the training of you data crosses the line into information you don't want to share until it's too late. It's also a slippery slope.