
Explore how capability-based systems like Sandstorm mitigate language-based security vulnerabilities through robust isolation and authorization mechanisms, enhancing web application security.

Explore verifiable computation for untrusted ASICs, enabling high-assurance execution without compromising performance. Learn about the Zebra system's implementation and its potential to match trusted platform performance.

Explores advanced techniques for mitigating code-reuse attacks at the binary level, focusing on improving Control-Flow Integrity and reducing potential targets for indirect branches.

Explore end-to-end integrity protection for web applications, focusing on safeguarding sensitive information against server-side attacks and ensuring data authenticity for users.

Explore how modern web technologies enable unique browser fingerprinting, its effectiveness on various devices, and potential privacy implications in the evolving digital landscape.

Investigates end-user behavior with found USB drives through a controlled experiment, revealing high success rates and quick responses, driven by altruistic intentions rather than technical incompetence.

Explore a usability-optimized decompiler for malware analysis, featuring readability improvements that significantly enhance analysts' ability to understand and combat complex malicious software.

Explore a novel language and system for secure, private gesture recognition in sensor-based interfaces, addressing privacy concerns and security threats while ensuring fast performance.

Explores secure oblivious storage systems for untrusted cloud providers, introducing TaoStore to address asynchronicity issues in multi-user scenarios with overlapping data sets.

Introduces a novel oblivious RAM supporting variable-sized blocks and a history-independent data structure, offering enhanced privacy and efficiency for key/value map operations.

Explore vulnerability scanning essentials for sysadmins, covering software vulnerabilities, networking, scanning tools, and server inspection. Learn how ops involvement can enhance overall security processes.

Explores users' perceptions of security and privacy in smart home assistants, revealing incomplete mental models and limited understanding of data handling, leading to inadequate threat models and coping strategies.

Explores vulnerabilities in AMD's Secure Encrypted Virtualization, demonstrating attacks on I/O operations and constructing primitives to breach VM memory encryption, highlighting security risks in emerging technology.

Explores bandwidth denial-of-service attacks against Tor, demonstrating feasibility and impact on network performance through experimentation and simulation. Analyzes costs and effects of various attack strategies.

Explores utility-optimized local differential privacy mechanisms for distribution estimation, enhancing data utility while maintaining privacy for sensitive information.

Explore the alarming rise of commercial spyware tools used by nation-states for covert surveillance and sabotage. Learn about Citizen Lab's methodology for uncovering abuses and the challenges posed by zero-click attacks.

Reverse engineering neural networks through electromagnetic side channels, revealing architecture details like activation functions, layers, neurons, and weights using non-invasive techniques on ARM Cortex-M3 microcontrollers.

Explore how embracing "hacky" solutions in cybersecurity can enhance enterprise defense. Learn innovative strategies and custom detection techniques to gain an advantage over attackers.

Explores alignment between COPPA regulations and parents' privacy expectations for smart toys, using contextual integrity surveys to quantify regulatory effectiveness and identify nuanced privacy concerns.

Investigates vulnerabilities in aircraft instrument landing systems to wireless attacks using software-defined radio, demonstrating potential for spoofing signals and causing off-runway landings in low-visibility conditions.

Unveils TESSERACT framework to eliminate spatial and temporal biases in Android malware classification, improving real-world robustness and revealing counterintuitive performance insights.