The opportunity may not be exactly what you'd think, because privacy is basically a fallacy for the kinds of information and images that are currently published

The opportunity may not be exactly what you'd think, because privacy is basically a fallacy for the kinds of information and images that are currently published. What has to change is the nature of the data that is public; specifically, data itself must become much less valuable, and this tends to require new protocols. For instance, you shouldn't have to "give" anything precious to a web site you barely know; there ought to be a way to hand them a proxy that becomes useless under certain conditions (e.g. after 24 hours have passed, after transferring to some other web site, or similar criteria). This reduces the security problem to one of probability: the time horizon is so small, and the data so fragile, that it is unlikely anyone will see a reason to steal it, or accidentally share it. Some solutions are available already. For example, people can create throw-away E-mail accounts to limit their exposure, and perhaps they just need to be encouraged to do so. Having said this, it's difficult to imagine ever securing some kinds of data. This is an age with nearly-ubiquitous mobile surveillance devices and Internet connectivity, so how can a web site even claim to safeguard images, for example? You can't know everyone who has seen you in public; you can't know about every cell phone picture that may have been taken of you, much less posted somewhere, with a "tag" showing your name. The best thing a web site can do is be honest about this reality, and flat out tell users that there is no way to perfectly-secure images of them.