SaTC: CORE: Small: Efficient Plausible Deniability Systems

Data privacy is essential in today’s computing-infused societal structures. Encryption is usually the first line of defense against privacy violations. However, upon compelled disclosure of encryption keys plaintexts are immediately revealed to powerful adversaries.

Oppressive regimes increasingly practice such forced disclosure to actively censor and filter networks, routinely investigate and imprison journalists and rights activists, and generally fight fiercely against any compromising information being presented to the outside world. The problem at the heart of this work is often a matter of life and death to journalists, human rights activists, whistleblowers, and individual citizens Existing systems that tries to address this problem, break down in the presence of realistic adversaries that can access the storage media at multiple points in time (“multi-snapshot” adversary).

This work will design and build strongly secure highly practical plausible deniable storage mechanisms that are robust against a multi-snapshot adversary enabling users to plausibly deny the possession of certain data even when the adversary has access to the user storage devices and credentials. Research results from this project would be incorporated into both undergraduate and graduate level courses and would help train the next generation workforce in plausible deniability paradigm and technology.

Plausible deniability is vital for human rights activists, whistleblowers in oppressive regimes, and citizens concerned about privacy as a right, since it is perhaps the only remaining recourse to ensuring personal privacy in a world of increasing control, oversight, and censorship.

This work will advance the foundational understanding of plausible deniability across a broad swath of storage systems and media. It will combine different technology, such as, faster access privacy mechanisms, data encoding mechanisms using write-once memory, trusted execution environments to harden deniability in the system stack and machine learning for addressing cover traffic information, to provide a robust plausibly deniable storage system.

It will build multiple systems that will be released for public use and integrated into mainstream operating systems. The work will also show the wide-ranging implications of strongly secure plausible deniability mechanisms by designing, building and demonstrating the world’s first “snoop-proof” Signal encrypted smartphone messaging application thereby defeating adversaries able to coerce users to reveal encryption keys.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.