Many cryptographic mechanisms depend upon the availability of securely generated random numbers. In practice, the sources of random numbers can be unreliable for many reasons, including bugs, compromise or subversion of standards. While there exist ways to significantly reduce the impact of unreliable randomness, these typically do not work well with practical constraints, such as long-term keys stored in hardware security modules. In practice, even modern protocols like TLS 1.3 lack such mechanisms and are therefore highly vulnerable to unreliable randomness. We propose a wrapper construction that reduces the impact of untrusted randomness, and which is is compatible with, and effective in, existing deployments of protocols such as TLS. We provide a security analysis of the construction and elaborate on design choices and practical interpretations. Our findings show that it is possible to effectively harden deployed protocols against unreliable randomness.

In the past few years secure messaging has become mainstream, with over a billion active users of end-to-end encryption protocols such as Signal. The Signal Protocol provides a strong property called post-compromise security to its users. However, it turns out that many of its implementations provide, without notification, a weaker property for group messaging: an adversary who compromises a single group member can read and inject messages indefinitely. We show for the first time that post-compromise security can be achieved in realistic, asynchronous group messaging systems. We present a design called Asynchronous Ratcheting Trees (ART), which uses tree-based Diffie-Hellman key exchange to allow a group of users to derive a shared symmetric key even if no two are ever online at the same time. ART scales to groups containing thousands of members, while still providing provable security guarantees. It has seen significant interest from industry, and forms the basis for two draft IETF RFCs and a chartered working group. Our results show that strong security guarantees for group messaging are practically achievable in a modern setting.

Signal is a new security protocol and accompanying app that provides end-to-end encryption for instant messaging. The core protocol has recently been adopted by WhatsApp, Facebook Messenger, and Google Allo among many others, the first two of these have at least 1 billion active users. Signal includes several uncommon security properties (such as "future secrecy" or "post-compromise security"), enabled by a novel technique called ratcheting in which session keys are updated with every message sent. Despite its importance and novelty, there has been little to no academic analysis of the Signal protocol. We conduct the first security analysis of Signal's key agreement and double ratchet as a multi-stage key exchange protocol. We extract from the implementation a formal description of the abstract protocol, and define a security model which can capture the "ratcheting" key update structure. We then prove the security of Signal's core in our model, demonstrating several standard security properties. We have found no major flaws in the design, and hope that our presentation and results can serve as a starting point for other analyses of this widely adopted protocol.

In this work we study communication with a party whose secrets have already been compromised. At first sight, it may seem impossible to provide any type of security in this scenario. However, under some conditions, practically relevant guarantees can still be achieved. We call such guarantees "post-compromise security". We provide the first informal and formal definitions for post-compromise security, and show that it can be achieved in several scenarios. At a technical level, we instantiate our informal definitions in the setting of authenticated key exchange (AKE) protocols, and develop two new strong security models for two different threat models. We show that both of these security models can be satisfied, by proposing two concrete protocol constructions and proving they are secure in the models. Our work leads to crucial insights on how post-compromise security can (and cannot) be achieved, paving the way for applications in other domains.