Adapt

Most cryptographic applications crucially rely on secret keys that are chosen randomly and are unknown to an attacker. Unfortunately, the process of deriving secret keys in practice is often difficult, error-prone and riddled with security vulnerabilities. Badly generated keys offer a prevalent source of attacks that render complex cryptographic applications completely insecure, despite their sophisticated design and rigorous mathematical analysis.

The computer security community has long advocated the concept of building multiple layers of defense to protect a system. Unfortunately, it has been difficult to realize this vision in the practice of software development, and software often ships with inadequate defenses, typically developed in an ad hoc fashion.

This project focuses on shoring up the security vulnerabilities that exist in computer processors. Just like in software, bugs in hardware present vulnerabilities that can be exploited by determined attackers. Prior work has developed a method whereby the processor monitors itself and sends an alert to software whenever dangerous, anomalous behavior is observed. The question of what constitutes dangerous behavior is an open one, and tackling it is the goal of this research.

This project studies the security of representative personalized services, such as search engines, news aggregators, and on-line targeted advertising, and identifies vulnerabilities in service components that can be exploited by pollution attacks to deliver contents intended by attackers.

This research is building an understanding of what data is useful to attackers and what data is private for its legitimate owners so that security systems can incorporate these values into a data-driven, defense-in-depth approach to securing our digital lives. We are exploiting the fact that both users and attackers must sift through vast amounts of data to find useful information.