integrating GPG with deniable steganography

[Florian Weimer]

Andrew Marlow <apm6674 at apmsoftwareltd.alkazar.co.uk> writes:

> Suppose Bob sends Alice a GPG-steg'd message. Wendy intercepts
> it and runs the decode program. She then serves Alice with
> an RIP decryption notice. Alice refuses. Her argument is that
> Wendy has recovered random data because the message she received
> was not concealing another message via the encode/decode programs.

I don't think this will work in practice.

Steganography is still in its infancy.  If the algorithms are public,
the party with the better noise model wins---and that's the government
agencies.

The only solution which seems to work at the moment is not hiding the
communication per se, but the amount of communication.  You might be
forced to reveal some parts of the message, but nobody can tell if
you've revealed all of it.  Of course, you might get into trouble
because they already know that there was additional information
transmitted (for example, because the other end revealed it), so this
scheme is most suitable for encryption of storage.

I think there's even a Linux file system which implements this
approach.  Have a look at Markus Kuhn's site.