[wirehead-wannabe]

Problems in privacy engineering that seem unsolvable:

• - sending information to another party that lets them observe and interact with it, but not store it indefinitely (or only lets them store it imperfectly)*
• - sending information to another party that lets them save it and interact with it however they want, but not share it with a third party*
• - verifying that one is not currently being observed (maybe use short-range EMPs to solve this in the case of checking a room for bugs?)*
• - being able to store and retrieve information from a device in a quickly and easily human-readabe format that no one else can understand
• - being able to e.g. enter passwords without anyone observing or understanding the step between thinking of the password in your mind and the device receiving the password*
• - encryption that can be broken only with a warrant somehow
• - being able to store information in such a way that it can be retrieved and used publicly, but not without the owner learned why and how you used it (this one may be very bad for people interested in reducing the power of IP laws)

Pretty sure many of these are actually theoretically impossible unless you can restrict the amount of surveillance or computational power that potential observers have access to.

The ones marked with a * are things that, as far as I can tell, intuitive social interaction and subjective feelings of security and privacy depend on. If they end up being major problems and sources of risk, I predict widespread mental health problems.

[mitigatedchaos]

Neural interface, brah. Helps with some of them.