Universal Circuits (UCs) offer a promising approach to hardware Intellectual Property (IP) obfuscation, leveraging cryptographic principles to hide both structure and function in a programmable logic fabric. Their adaptability makes them especially suitable for the globalized Integrated Circuit (IC) supply chain, where security against threats like reverse engineering is crucial. Despite the potential, UC security remains largely unexplored. This work evaluates UC security against state-of-the-art oracle-guided (OG) and oracle-less (OL) attacks. Results show near-random success rates (approx 50

Recent advances in logic locking have produced numerous schemes, many of which have been evaluated using informal methods, leaving them vulnerable to attacks. While formal frameworks for logic locking, such as universal circuits (UC), offer security against reverse engineering, their practical implementation suffers from significant overhead costs. In this work, we introduce Universal Topological Arrays, a novel, scalable alternative to traditional UC and embedded FPGA (eFPGA) designs. Our approach leverages a programmable, regular array architecture that enables efficient topological mapping of circuits while minimizing overhead. By reducing the critical path length and eliminating unnecessary supplementary circuitry, our solution addresses key limitations of existing UC and eFPGA-based methods. While also programmable, our design achieves over 10x improvement in energy efficiency compared to UC implementations, while being 2.3x smaller than minimum-sized eFPGAs. Additionally, it scales effectively with increasing circuit sizes, representing a significant step toward practical, scalable, and provably secure hardware solutions for logic locking.

Growing IC manufacturing complexity and reliance on third-party fabrication create supply chain fragility, contributing to chip shortages and IP security risks. General-purpose ICs can mitigate manufacturing security risks but rely on software-based configurations, which is not optimal for high-consequence applications. Our work proposes a novel IP-agnostic Foundational Cell Array (FC-Array) platform to overcome these challenges. Built on verified standard cells and industry-standard EDA tools, this platform preserves many advantages of an ASIC. By incorporating 3D split manufacturing, we provide semantically secure IP protection and a base wafer that can be stockpiled. Our tests demonstrate both power-efficient (100 MHz) and high-performance (1 GHz) options. In a post-place-and-route simulated 28nm design, our FC-Array shows a worst-case 1.85x increase in power consumption and a 2.61x increase in area compared to standard cell ASICs for equivalent timing performance.

The outsourced manufacturing of integrated circuits has increased the risk of intellectual property theft. In response, logic locking techniques have been developed for protecting designs by adding programmable elements to the circuit. These techniques differ significantly in both overhead and resistance to various attacks, leaving designers unable to discern their efficacy. To overcome this critical impediment for the adoption of logic locking, we propose two metrics, key corruption and minimum corruption, that capture the goals of locking under different attack scenarios. We develop a flow for approximating these metrics on generic locked circuits and evaluate several locking techniques.

Modern commercial antivirus systems increasingly rely on machine learning (ML) to keep up with the rampant inflation of new malware. However, it is well-known that machine learning models are vulnerable to adversarial examples (AEs). Previous works have shown that ML malware classifiers are fragile to the white-box adversarial attacks. However, ML models used in commercial antivirus (AV) products are usually not available to attackers and only return hard classification labels. Therefore, it is more practical to evaluate the robustness of ML models and real-world AVs in a pure black-box manner. We propose a black-box Reinforcement Learning (RL) based framework to generate AEs for PE malware classifiers and AV engines. It regards the adversarial attack problem as a multi-armed bandit problem, which finds an optimal balance between exploiting the successful patterns and exploring more varieties. Compared to other frameworks, our improvements lie in three points: 1) limiting the exploration space by modeling the generation process as a stateless process to avoid combination explosions, 2) reusing the successful payload in modeling; and 3) minimizing the changes on AE samples to correctly assign the rewards in RL learning (which also helps identify the root cause of evasions). As a result, our framework has much higher evasion rates than other off-the-shelf frameworks. Results show it has over 74%--97% evasion rate for two state-of-the-art ML detectors and over 32%--48% evasion rate for commercial AVs in a pure black-box setting. We also demonstrate that the transferability of adversarial attacks among ML-based classifiers is higher than that between ML-based classifiers and commercial AVs.

Modern commercial antivirus systems increasingly rely on machine learning (ML) to keep up with the rampant inflation of new malware. However, it is well-known that machine learning models are vulnerable to adversarial examples (AEs). Previous works have shown that ML malware classifiers are fragile to the white-box adversarial attacks. However, ML models used in commercial antivirus (AV) products are usually not available to attackers and only return hard classification labels. Therefore, it is more practical to evaluate the robustness of ML models and real-world AVs in a pure black-box manner. We propose a black-box Reinforcement Learning (RL) based framework to generate AEs for PE malware classifiers and AV engines. It regards the adversarial attack problem as a multi-armed bandit problem, which finds an optimal balance between exploiting the successful patterns and exploring more varieties. Compared to other frameworks, our improvements lie in three points: 1) limiting the exploration space by modeling the generation process as a stateless process to avoid combination explosions, 2) reusing the successful payload in modeling; and 3) minimizing the changes on AE samples to correctly assign the rewards in RL learning (which also helps identify the root cause of evasions). As a result, our framework has much higher evasion rates than other off-the-shelf frameworks. Results show it has over 74%-97% evasion rate for two state-of-the-art ML detectors and over 32%-48% evasion rate for commercial AVs in a pure black-box setting. We also demonstrate that the transferability of adversarial attacks among ML-based classifiers is higher than that between ML-based classifiers and commercial AVs.

Modern commercial antivirus systems increasingly rely on machine learning to keep up with the rampant inflation of new malware. However, it is well-known that machine learning models are vulnerable to adversarial examples (AEs). Previous works have shown that ML malware classifiers are fragile to the white-box adversarial attacks. However, ML models used in commercial antivirus products are usually not available to attackers and only return hard classification labels. Therefore, it is more practical to evaluate the robustness of ML models and real-world AVs in a pure black-box manner. We propose a black-box Reinforcement Learning (RL) based framework to generate AEs for PE malware classifiers and AV engines. It regards the adversarial attack problem as a multi-armed bandit problem, which finds an optimal balance between exploiting the successful patterns and exploring more varieties. Compared to other frameworks, our improvements lie in three points. 1) Limiting the exploration space by modeling the generation process as a stateless process to avoid combination explosions. 2) Due to the critical role of payload in AE generation, we design to reuse the successful payload in modeling. 3) Minimizing the changes on AE samples to correctly assign the rewards in RL learning. It also helps identify the root cause of evasions. As a result, our framework has much higher black-box evasion rates than other off-the-shelf frameworks. Results show it has over 74\%--97\% evasion rate for two state-of-the-art ML detectors and over 32\%--48\% evasion rate for commercial AVs in a pure black-box setting. We also demonstrate that the transferability of adversarial attacks among ML-based classifiers is higher than the attack transferability between purely ML-based and commercial AVs.

In this paper, we provide the first large-scale empirical analysis of IoT devices in real-world homes by leveraging data collected from user-initiated network scans of 83M devices in 16M households. We find that IoT adoption is widespread: on several continents, more than half of households already have at least one IoT device. Device types and manufacturer popularity vary dramatically across regions. For example, while nearly half of North American homes have an Internet-connected television or streaming device, less than three percent do in South Asia where the majority of devices are surveillance cameras. We investigate the security posture of devices, detailing their open services, weak default credentials, and vulnerability to known attacks. Device security similarly varies geographically, even for specific manufacturers. For example, while less than 17% of TP-Link home routers in North America have guessable passwords, nearly half do in Eastern Europe and Central Asia. We argue that IoT devices are here, but for most homes, the types of devices adopted are not the ones actively discussed. We hope that by shedding light on this complex ecosystem, we help the security community develop solutions that are applicable to today's homes.