Dark patterns have gained increasing attention among the HCI and design communities, but little is known about how they intersect with assistive technologies (ATs) and impact people with accessibility needs, such as blind and low-vision (BLV) individuals. To address this gap, we conducted an in-lab user study with 18 BLV participants using a custom-built social media application that embeds six common dark patterns. Through observing participant experiences with assigned tasks and semi-structured post-study interviews, we explored how screen readers and magnification tools influence the perception and amplification of deceptive design elements. In contrast to prior work that identified accessibility-induced deception, our findings demonstrate a dual role of ATs where dark patterns are either revealed or intensified. Screen readers exposed hidden manipulations like bad defaults but amplified other dark patterns through sequential reading. Similarly, magnifiers intensified deceptive effects through viewport reduction by restricting the visible area. We conceptualize this mechanism as assistive amplification and show how dark patterns manifest differently for BLV users, informing the design of more inclusive and manipulation-resistant interfaces.

Traffic watermarking is a powerful traffic-analysis technique and remains a practical threat to anonymous communication systems, yet its real-world feasibility is still poorly understood. Prior schemes are typically evaluated in narrow, single-scenario setups with ad hoc implementations and implicit assumptions about flow structure, coordination between embedder and detector, and how perturbations survive network noise. This makes results hard to reproduce, compare across contexts, and stress-test against modern network protocol features. This paper introduces Waterfall, a capsule-based framework for implementing and evaluating watermarking techniques on real traffic. Waterfall captures common semantic foundations of watermarking and exposes them via a unified capsule abstraction with a programmable API and declarative configuration. We implement Waterfall and validate it by reproducing and stress-testing representative schemes from the literature, and by deriving self-adaptive variants that tune their behavior using live traffic measurements. We further use Waterfall to analyze Tor's deployed Conflux protocol and show that its traffic-splitting design can unintentionally amplify watermark detectability. Finally, we demonstrate that Waterfall is also suitable for implementing and evaluating defenses against watermarking attacks.

This paper provides a comprehensive guide for conducting penetration tests in fifth generation (5G) networks, particularly in campus environments, to enhance security of these networks. While 5G technology advances areas such as the Internet of Things (IoT), autonomous systems, and smart cities, its complex, virtualized, and open architecture also introduces new security risks. The paper outlines methods for identifying vulnerabilities in key 5G components, including the Radio Access Network (RAN), Core Network, and User Equipment (UE), to address emerging threats such as protocol manipulation or user tracking. This paper analyzes the current scientific literature and evaluates whether attacks can be used in a penetration-testing scenario. We identify current attacks and tools and consider them multidimensional regarding STRIDE threats and violations of the security dimensions. We release an extended version of MITRE Enterprise ATT&CK that contains our identified data.

While protecting user data is essential, software developers often fail to fulfill privacy requirements. However, the reasons why they struggle with privacy-compliant implementation remain unclear. Is it due to a lack of knowledge, or is it because of insufficient support? To provide foundational insights in this field, we conducted a qualitative 5-hour programming study with 30 professional software developers implementing 3 privacy-sensitive programming tasks that were designed with GDPR compliance in mind. To explore if and how developers implement privacy requirements, participants were divided into 3 groups: control, privacy prompted, and privacy expert-supported. After task completion, we conducted follow-up interviews. Alarmingly, almost all participants submitted non-GDPR-compliant solutions (79/90). In particular, none of the 3 tasks were solved privacy-compliant by all 30 participants, with the non-prompted group having the lowest number of 3 out of 30 privacy-compliant solution attempts. Privacy prompting and expert support only slightly improved participants' submissions, with 6/30 and 8/30 privacy-compliant attempts, respectively. In fact, all participants reported severe issues addressing common privacy requirements such as purpose limitation, user consent, or data minimization. Counterintuitively, although most developers exhibited minimal confidence in their solutions, they rarely sought online assistance or contacted the privacy expert, with only 4 out of 10 expert-supported participants explicitly asking for compliance confirmation. Instead, participants often relied on existing implementations and focused on implementing functionality and security first.

Current approaches for emulating cellular base-bands inherently fall short in comparison to over-the-air testing due to their limited support for the complex peripherals involved in a modern baseband, such as DSPs, SIM cards and RF frontends. Improving such support is a daunting task, requiring deep reverse-engineering which is extremely time consuming - resulting in slow progress. Consequently, techniques such as fuzzing are only able to find relatively shallow bugs, since they are unable to reach the states required for the majority of the baseband to function. To fill this gap, we propose Basebridge, which enables far more comprehensive simulation of baseband behavior by restoring relevant state from memory dumps of real devices. Our prototype implementation supports baseband firmware from two major vendors (MediaTek and Samsung), and - in contrast to current state-of-the-art emulators - correctly responds to 97% of tested RRC and NAS messages while improving coverage by an average factor of 2.41 (Samsung) and 5.54 (MediaTek). Basebridge also passes several LTE conformance tests. Our empirical evaluation demonstrates that this enhanced fidelity enables faster discovery of a wider range of bugs thanks to the scalability of emulation; our fuzzing campaign shows that coverage improves by a factor of 2.3-5x overall, and by a factor of 9.0-22.5x for functionality targeted by our approach. Basebridge unveiled 5 new vulnerabilities, which we have disclosed to affected vendors.

The introduction of the General Data Protection Regulation (GDPR) in 2018 marked a pivotal moment in the evolution of data protection within the European Union (EU). Consequently, companies have since been legally obliged to respect users' privacy, and, if found to be in violation, risk incurring fines. While this regulatory change greatly benefits users, software developers, on the other hand, face a tremendous challenge to make their applications compliant, creating a gap between legal requirements and effective software development. Several solutions have been proposed to bridge the gap for web application developers. However, it is unclear to what extent they fulfill the requirements laid out by the GDPR. In this work, we look at three frameworks that aim to aid compliance for web applications. To efficiently assess them, we propose a methodology and several benchmarks to evaluate and compare the frameworks. From the GDPR, we have derived a set of requirements that do not entail institutional changes but have technical implications for software. Leveraging these requirements, we evaluate both the proposed solutions' enforcement capabilities and computational overhead. Our comparison shows that each framework can, if configured correctly, enforce a different subset of GDPR requirements. Finally, based on the insights gained, we provide recommendations for the community on how to make further progress on operationalizing the GDPR.

Android operating system restricts access to data by enabling data control flow and permission systems to reduce the risk of information theft. Therefore, attackers are constantly looking for alternative and stealthy approaches to exfiltrate private data from a targeted device. This paper presents CovertPower, a covert channel attack that exfiltrates user data by actively inducing power consumption on Android devices. At the transmitting end, our CovertPower app modulates binary data into a timed resource workload (e.g., processor, write-on-memory), producing power consumption bursts. On the receiving end, we acquire power consumption traces via a low-cost hardware tool that can be easily concealed in USB wall-socket adapters or powerbanks. Therefore, a signal processing-based decoder analyzes such traces and retrieves the exfiltrated information. We demonstrate the feasibility of our attack with a thorough experimental evaluation on 14 mobile devices and various real-world settings such as display state, ongoing activities, and charging technologies. Our attack achieves a transfer speed of up to 10 bps with a high bit sequence similarity on most devices and settings considered.

Machine Learning (ML) promises to enhance the efficacy of Android Malware Detection (AMD); however, ML models are vulnerable to realistic evasion attacks—crafting realizable Adversarial Examples (AEs) that satisfy Android malware domain constraints. To eliminate ML vulnerabilities, defenders aim to identify susceptible regions in the feature space where ML models are prone to deception. The primary approach to identifying vulnerable regions involves investigating realizable AEs, but generating these feasible apps poses a challenge. For instance, previous work has relied on generating either feature-space norm-bounded AEs or problem-space realizable AEs in adversarial hardening. The former is efficient but lacks full coverage of vulnerable regions, whereas the latter can uncover these regions by satisfying domain constraints but is known to be time consuming. To address these limitations, we propose an approach to facilitate the identification of vulnerable regions. Specifically, we introduce a new interpretation of Android domain constraints in the feature space, followed by a novel technique that learns them. Our empirical evaluations across various evasion attacks indicate effective detection of AEs using learned domain constraints, with an average of 89.6%. Furthermore, extensive experiments on different Android malware detectors demonstrate that utilizing our learned domain constraints in adversarial training outperforms other adversarial training based defenses that rely on norm-bounded AEs or state-of-the-art non-uniform perturbations. Finally, we show that retraining a malware detector with a wide variety of feature-space realizable AEs results in a 77.9% robustness improvement against realizable AEs generated by unknown problem-space transformations, with up to 70× faster training than using problem-space realizable AEs.

On the modern web, trackers and advertisers frequently construct and monetize users’ detailed behavioral profiles without consent. Despite various studies on web tracking mechanisms and advertisements, there has been no rigorous study focusing on websites targeted at children. To address this gap, we present a measurement of tracking and (targeted) advertising on websites directed at children. Motivated by the lack of a comprehensive list of child-directed (i.e., targeted at children) websites, we first build a multilingual classifier based on web page titles and descriptions. Applying this classifier to over two million pages from the Common Crawl dataset, we compile a list of two thousand child-directed websites. Crawling these sites from five vantage points, we measure the prevalence of trackers, fingerprinting scripts, and advertisements. Our crawler detects ads displayed on child-directed websites and determines if ad targeting is enabled by scraping ad disclosure pages whenever available. Our results show that around 90% of child-directed websites embed one or more trackers, and about 27% contain targeted advertisements—a practice that should require verifiable parental consent. Next, we identify improper ads on child-directed websites by developing an ML pipeline that processes both images and text extracted from ads. The pipeline allows us to run semantic similarity queries for arbitrary search terms, revealing ads that promote services related to dating, weight loss, and mental health, as well as ads for sex toys and flirting chat services. Some of these ads feature repulsive, sexually-explicit and highly-inappropriate imagery. In summary, our findings indicate a trend of non-compliance with privacy regulations and troubling ad safety practices among many advertisers and child-directed websites. To ensure the protection of children and create a safer online environment, regulators and stakeholders must adopt and enforce more stringent measures. Keywords – online tracking, advertising, children, privacy

Over the last decade, researchers have extensively explored the vulnerabilities of Android malware detectors to adversarial examples through the development of evasion attacks; however, the practicality of these attacks in real-world scenarios remains arguable. The majority of studies have assumed attackers know the details of the target classifiers used for malware detection, while in reality, malicious actors have limited access to the target classifiers. This paper introduces EvadeDroid, a problem-space adversarial attack designed to effectively evade black-box Android malware detectors in real-world scenarios. EvadeDroid constructs a collection of problem-space transformations derived from benign donors that share opcode-level similarity with malware apps by leveraging an n-gram-based approach. These transformations are then used to morph malware instances into benign ones via an iterative and incremental manipulation strategy. The proposed manipulation technique is a query-efficient optimization algorithm that can find and inject optimal sequences of transformations into malware apps. Our empirical evaluations, carried out on 1 K malware apps, demonstrate the effectiveness of our approach in generating real-world adversarial examples in both soft- and hard-label settings. Our findings reveal that EvadeDroid can effectively deceive diverse malware detectors that utilize different features with various feature types. Specifically, EvadeDroid achieves evasion rates of 80%-95% against DREBIN, Sec-SVM, ADE-MA, MaMaDroid, and Opcode-SVM with only 1-9 queries. Furthermore, we show that the proposed problem-space adversarial attack is able to preserve its stealthiness against five popular commercial antiviruses with an average of 79% evasion rate, thus demonstrating its feasibility in the real world.

To ensure data-privacy compliance, it is common for companies to consult privacy experts for the identification and communication of privacy requirements to software developers. However, developers often fail to fulfill those requirements resulting in companies regularly being fined for violations due to non-compliance with privacy data regulations. To investigate why software developers struggle with the implementation of privacy requirements and explore their communication modality, we conducted a qualitative semi-structured interview study with 30 participants involving 10 software developers, 10 privacy experts, and 10 team coordinators with an average experience of nine years in the privacy communication and implementation process within a company context. We found a communication gap between software developers and privacy experts, suggesting a lack of proper procedural steps during the software development process to guarantee that the privacy requirements have been adequately addressed. We also uncovered that since privacy requirements were mostly communicated in a uni-directional manner, they were often perceived as a hindrance during software development, thus fostering an adversarial relationship between privacy experts and developers. Therefore, in order to fulfill the experts' requirements, software developers requested concrete steps to take during the software development process, as observed in the security field. However, privacy experts often lacked the technical knowledge to provide such instructions. This work contributes an explanatory theory on the communication gap between software developers and privacy experts. We discuss common obstacles in the communication of privacy experts and software developers and provide guidance on how to address them.

Wireless radio channels are known to contain information about the surrounding propagation environment, which can be extracted using established wireless sensing methods. Thus, today’s ubiquitous wireless devices are attractive targets for passive eavesdroppers to launch reconnaissance attacks. In particular, by overhearing standard communication signals, eavesdroppers obtain estimations of wireless channels which can give away sensitive information about indoor environments. For instance, by applying simple statistical methods, adversaries can infer human motion from wireless channel observations, allowing to remotely monitor premises of victims. In this work, building on the advent of intelligent reflecting surfaces (IRSs), we propose IRShield as a novel countermeasure against adversarial wireless sensing. IRShield is designed as a plug-and-play privacy-preserving extension to existing wireless networks. At the core of IRShield, we design an IRS configuration algorithm to obfuscate wireless channels. We validate the effectiveness with extensive experimental evaluations. In a state-of-the-art human motion detection attack using off-the-shelf Wi-Fi devices, IRShield lowered detection rates to 5% or less.

Abstract In this paper we present LiM (‘Less is More’), a malware classification framework that leverages Federated Learning to detect and classify malicious apps in a privacy-respecting manner. Information about newly installed apps is kept locally on users’ devices, so that the provider cannot infer which apps were installed by users. At the same time, input from all users is taken into account in the federated learning process and they all benefit from better classification performance. A key challenge of this setting is that users do not have access to the ground truth (i.e. they cannot correctly identify whether an app is malicious). To tackle this, LiM uses a safe semi-supervised ensemble that maximizes classification accuracy with respect to a baseline classifier trained by the service provider (i.e. the cloud). We implement LiM and show that the cloud server has F1 score of 95%, while clients have perfect recall with only 1 false positive in > 100 apps, using a dataset of 25K clean apps and 25K malicious apps, 200 users and 50 rounds of federation. Furthermore, we conduct a security analysis and demonstrate that LiM is robust against both poisoning attacks by adversaries who control half of the clients, and inference attacks performed by an honest-but-curious cloud server. Further experiments with Ma-MaDroid’s dataset confirm resistance against poisoning attacks and a performance improvement due to the federation.

A wave of alternative coins that can be effectively mined without specialized hardware, and a surge in cryptocurrencies' market value has led to the development of cryptocurrency mining ( cryptomining ) services, such as Coinhive, which can be easily integrated into websites to monetize the computational power of their visitors. While legitimate website operators are exploring these services as an alternative to advertisements, they have also drawn the attention of cybercriminals: drive-by mining (also known as cryptojacking ) is a new web-based attack, in which an infected website secretly executes JavaScript code and/or a WebAssembly module in the user's browser to mine cryptocurrencies without her consent. In this paper, we perform a comprehensive analysis on Alexa's Top 1 Million websites to shed light on the prevalence and profitability of this attack. We study the websites affected by drive-by mining to understand the techniques being used to evade detection, and the latest web technologies being exploited to efficiently mine cryptocurrency. As a result of our study, which covers 28 Coinhive-like services that are widely being used by drive-by mining websites, we identified 20 active cryptomining campaigns. Motivated by our findings, we investigate possible countermeasures against this type of attack. We discuss how current blacklisting approaches and heuristics based on CPU usage are insufficient, and present MineSweeper, a novel detection technique that is based on the intrinsic characteristics of cryptomining code, and, thus, is resilient to obfuscation. Our approach could be integrated into browsers to warn users about silent cryptomining when visiting websites that do not ask for their consent.

Tracking services play a fundamental role in the smartphone ecosystem. While their primary purpose is to provide a smartphone user with the ability to regulate the extent of sharing private information with external parties, these services can also be misused by advertisers in order to boost revenues. In this paper, we investigate tracking services on the Android and iOS smartphone platforms. We present a simple and effective way to monitor traffic generated by tracking services to and from the smartphone and external servers. To evaluate our work, we dynamically execute a set of Android and iOS applications, collected from their respective official markets. Our empirical results indicate that even if the user disables or limits tracking services on the smartphone, applications can by-pass those settings and, consequently, leak private information to external parties. On the other hand, when testing the location 'on' setting, we notice that generally location is not tracked.