Natural Language Steganography and an “AI-complete” Security Primitive

In order to sketch, why steganography is such an important topic, and has received far too little attention from the hacker community in the past, let me quickly challenge our view of cryptosystems as commonly built in the context of military or commercial applications: Cryptosystems are designed to protect our sensitive data from evil arbitrators. Wrong. Well, maybe not. But then again, what is evil? Basically one could say, hacker ethics is about protecting a good individual from a bad society and this has some severe consequences in the construction of secure communication systems. In particular, I would like to draw our attention to the evil spy, we are envisioning, when we think about cryptosystems, who intercepts sensitive military communication, and to the criminal who fakes banking transactions. Most often, what we have in mind, when we build cryptosystems is a bad individual in a good society, which is rather some instance of witch hunt ethics. In practice this assumption has the simple incarnation, that cryptograms are vulnerable to detection, since a good society would encourage the use of cryptography as a means to protect everyone’s privacy. But under the assumption of a bad society, would Alice and Bob be allowed to use cryptography? What would witchhunt ethics assert about people who use cryptography? Do they have something to hide? Something evil? The central shift in views that is necessary, is the fact that it is not Alice and Bob who control their communication channel, so they can make sure, that evil Wendy won’t be able to recover their communication, but evil Wendy controls the channel and wants to make sure Alice and Bob can’t exchange unwanted messages. This communication setup has first been stated by Gustavus J. Simmons, which he popularly introduced via the prisoner’s dilemma. In this scenario we assume Alice and Bob are arrested, and “their only means of communication after they are locked up will be by way of messages conveyed for them by trustees – who are known to be agents of the warden... However since he has every reason to suspect that the prisoners want to coordinate an escape plan, the warden will only permit the exchanges to occur if the information contained in the messages is completely open to him – and presumably innocuous.” (Simmons 1984).